Browsing by Subject "Microelectromechanical systems"
Now showing 1 - 20 of 20
Results Per Page
Sort Options
Item A finite element test bed for development of feedback control laws for electrostatic MEMS(2005-12) Kawade, Balasaheb D.; Berg, Jordan M.; Dallas, Timothy E. J.; Idesman, Alexander V.This project presents the ANSYS simulation techniques for an electrostatically-actuated MEMS device incorporating feedback control laws. The electrostatic MEMS device consists of a movable electrode, suspended on flexible, elastic structures, and one or more fixed drive electrodes. Nonlinear feedback control laws are simulated in ANSYS multi-physics solver and a transducer element. ANSYS multi-physics solver is limited for these types of simulations. ANSYS doesn’t support multiframe restart and the combined circuit and electrostatic analysis are incompatible. This work presents simulation techniques based on numerical methods to circumvent these limitations. The proposed technique eliminates the circuit elements from the model, and instead propagates the associated states in an APDL macro. ANSYS auto time stepping method is not applicable for closed-loop feedback control systems because loads are calculated at each step based on simulation output at the previous step. An adaptive step size Runge-Kutta integration routine is incorporated within APDL macro to develop an efficient simulation technique. The simulation efficiency of the static closed loop feedback control systems is increased by a factor more than 100. However, a dynamic closed loop feedback control systems exhibits only a brief initial transient, and then does not permit further step size increases. To increase the simulation efficiency of such systems, the adaptation logic is turned off once the step size stabilizes. Simulation results for representative MEMS devices including a one-DOF piston microactuator and a two-DOF rotating/translating microactuator demonstrate the efficiency of these simulation techniques.Item A finite element test bed for development of feedback control laws for electrostatic MEMS(Texas Tech University, 2005-12) Kawade, Balasaheb D.; Berg, Jordan M.; Dallas, Timothy E. J.; Idesman, Alexander V.This project presents the ANSYS simulation techniques for an electrostatically-actuated MEMS device incorporating feedback control laws. The electrostatic MEMS device consists of a movable electrode, suspended on flexible, elastic structures, and one or more fixed drive electrodes. Nonlinear feedback control laws are simulated in ANSYS multi-physics solver and a transducer element. ANSYS multi-physics solver is limited for these types of simulations. ANSYS doesn’t support multiframe restart and the combined circuit and electrostatic analysis are incompatible. This work presents simulation techniques based on numerical methods to circumvent these limitations. The proposed technique eliminates the circuit elements from the model, and instead propagates the associated states in an APDL macro. ANSYS auto time stepping method is not applicable for closed-loop feedback control systems because loads are calculated at each step based on simulation output at the previous step. An adaptive step size Runge-Kutta integration routine is incorporated within APDL macro to develop an efficient simulation technique. The simulation efficiency of the static closed loop feedback control systems is increased by a factor more than 100. However, a dynamic closed loop feedback control systems exhibits only a brief initial transient, and then does not permit further step size increases. To increase the simulation efficiency of such systems, the adaptation logic is turned off once the step size stabilizes. Simulation results for representative MEMS devices including a one-DOF piston microactuator and a two-DOF rotating/translating microactuator demonstrate the efficiency of these simulation techniques.Item Analysis of stress singularity of adhered contacts in MEMS(Texas Tech University, 2004-08) Chakkarapani, VenkatasubbaraoMEMS devices are usually multimaterial systems where interfaces are formed at the junction of two materials. Failure occurs at adhered contacts because of biomaterial stress singularities at interface comers. Magnitude of the stress field induced due to this singularity is given by the value of the notch stress intensity. Hence it becomes very important to design MEMS devices based on the stress intensity-fracture toughness failure criterion. Inherent uncertainty of design parameters (which includes singularity parameters) in MEMS devices necessitates probabilistic design rather than deterministic design. The probabilistic design of MEMS devices, with a microswitch as our device example, has been performed to find the probability of failure of the switch based on stress intensity-fracture toughness failure criterion. The two main objectives of this research are to determine the stress field around a bimaterial singularity for a given bimaterial specimen and evaluate the probability of failure based on stress intensity-fracture toughness failure criterion using probabilistic analysis. The scope of work is fourfold. First, the order of the singularity is determined using two different methods, namely, Complex potential method and Airy stress function method. The equivalence of these methods is verified. Second, the influence coefficients are determined using analytical methods. Third, the stress intensity factor is determined using finite element methods. Fourth, the probabilistic analysis of the microswitch is performed based on stress intensity-fracture toughness failure criterion. The order of singularity has been determined to be 0.512 and 0.696. The stress intensity factor has been determined to be 0.7708 MPa m0.488 from finite element analysis. The probability that the notch stress intensity exceeds the fracture toughness is found to be 0.612.Item Design and fabrication of a MEMS micromirror with integrated charge sensor for feedback control(Texas Tech University, 2004-12) Anderson, Robert ChristopherThe "snap-through" or "pull-in" phenomenon limits the operating region of an electrostatic MEMS device controlled using a constant bias voltage to only one third of its physical deflection range. It has been shown that by controlling the charge on the actuator any static deflection inside the full gap range can theoretically be stabilized [18, 21, 22, 23]. Some existing approaches to this problem use passive charge feedback, implemented, for example, through a series capacitor. Others use open-loop charge control, involving the application of charge to a fixed capacitor, and the transfer of that charge to the MEMS. Electrostatically-actuated MEMS structures are fabricated at Texas Tech University using surface micromachining methods. The devices are intended to test voltage control laws now under development. The test structure includes the auxiliary electrode located directly on top of the lower drive electrode, to establish an integrated capacitor in order to make the desired charge measurement. This thesis concentrates on the manufacturability of such a device. Several fabrication processes were developed specifically for the previously mentioned devices. These processes include an isotropic plasma etch for deep undercut of MEMS structures, a low stress metal deposition process, and a photoresist profile modification procedure to ensure conformal step coverage in an electron beam metal deposition system. The development of several mirror designs are discussed, leading to the successful fabrication of a MEMS micromirror with integrated charge sensing capabilities.Item Dynamics of an electrostatically controlled bulk micromachined silicon torsion mirror in air(Texas Tech University, 2004-08) Gokarnesan, ManikandanStatic and dynamic characterizations of micromirrors are important to understand the mirror's responses for different inputs under various conditions. The characterizations are important not only to choose and operate a mirror for a particular end application but also to continuously provide feedback to the design and processing teams in a new mirror development. This feedback should include device behavior, system parameters, and material properties. The dynamic behavior of the mirrors is significant to implement adaptive control algorithms to precisely position the mirror. In this thesis, instrumentation techniques to measure the static and dynamic responses of a bulk micromachined Silicon torsion mirror have been developed. These measurements include tilt angle versus driving voltage curves and frequency response curves. Several important parameters, like the torsion spring constant, damping ratio, natural frequency, moment of inertia, and mass were estimated for the mirror. These parameters were incorporated in the mirror model to simulate the mirror responses. Agreement of the simulated and experimental results confirmed the validity of the measurement techniques. Effects of nonlinearity in the torsion springs resulted in deviations between the actual and the simulated mirror responses at higher voltages and at larger angles. The measured parameters were further used to simulate the dynamic responses, response to a step voltage, capacitance variations, and the dynamic deformations of the mirror. The various parameters estimated will be central to implementing feedback control algorithms to accurately position the mirror throughout the entire gap.Item Experimental study of fast electrons from the interaction of ultra intense laser and solid density plasmas(2008-08) Cho, Byoung-ick, 1976-; Ditmire, Todd R.A series of experiments have been performed to understand fast electron generation from ultra intense laser-solid interaction, and their transports through a cold material. Using Micro-Electro-Mechanical Systems (MEMS), we contrived various shape of cone and wedge targets. The first set of experiment was for investigating hot electron generations by measuring x-ray production in different energy ranges. K[alpha] and hard x-ray yields were compared when the laser was focused into pyramidal shaped cone targets and wedge shaped targets. Hot electron production is highest in the wedge targets irradiated with transverse polarization, though K[alpha] is maximized with wedge targets and parallel polarization. These results are explained with particle-in-cell (PIC) simulations utilizing PICLS and OOPIC codes. We also investigate hot electron transport in foil, wedge, and cone targets by observing the transition radiation emitted from the targets rear side along with bremsstrahlung x-ray measurement. Twodimensional images and spectra of 800 nm coherent transition radiation (CTR) along with ballistic electron transport analysis have revealed the spatial, temporal, and temperature characteristics of hot electron micro-pulses. Various patterns from different target-laser configurations suggest that hot electrons were guided by the strong static electromagnetic fields at the target boundary. Evidence about fast electron guiding in the cone is also observed. CTR at 400 nm showed that two distinct beams of MeV electrons are emitted from the target rear side at the same time. This measurement indicates that two different mechanisms, namely resonance absorption and j x B heating, create two populations of electrons at the targets front side and drive them to different directions, with distinct temperatures and temporal characteristics. This interpretation is consistent with the results from 3D-PIC code Virtual Laser Plasma Laboratory (VLPL).Item Extremes in the distribution of micro mechanical properties in a micro electrical mechanical device(2011-05) Patterson, Timothy L.; Gale, Richard O.; Dallas, Timothy E. J.Texas Instruments’ DLP group forever changed projection technology and the widespread adoption of projection systems with the creation of the digital micromirror device. The DMD can now be found enabling projection systems for many different applications, from movie theatres to classrooms, all across the world. Due to the unique characteristics of the DMD and demanding applications in which it is used, the DLP group has devised equally unique testing methods for ensuring robust and desirable operation of the DMD that meets the high standards expected in today’s technology. As production volumes dramatically increase with smaller, more cost effective devices, the challenge of ensuring optimal use of testing resources has once again become a point of focus for the DPL group. The investigation outlined in this thesis is based on the motivation of optimizing testing efficiency through the development of new testing methodologies based on results from modeling the behavior of components within the DMD array.Item Iterative approach to deformable membrane design(Texas Tech University, 2004-05) Nellore, Prashanth RA continuous deformable membrane mirror can be fabricated using MEMS processes. It will have a highly reflective thin membrane, which is the mirror surface, and a number of actuators below it. The shape of the mirror surface is controlled by inducing an electrostatic force through applying a voltage between the mirror surface and the actuators. Small displacements of a membrane under tension are governed by Poisson's equation. The solutions of this equation determine the deflections in the vertical direction for different pressures applied at various surface points, due to various electrode positions. This thesis examines the influence of number, position, thickness of the electrodes and also the electrostatic pressure applied by each electrode on the surface of the mirror. A Matlab Graphical User Interface is designed that computes the deflections and uses iterative methods to optimize the voltages on the electrodes, their position and thickness. It produces a deformable membrane design that will match the surface of the mirror to that of a spherical mirror of known radius of curvature.Item Mems based bead size selection method for the electronic taste chip(2004) Park, Byunghwa; Neikirk, Dean P.A micromachined biological and chemical sensor array has been developed for the rapid characterization of multiple analytes in solution. Various biochemical and chemical sensors are loaded in small micromachined structure together to analyze the ingredients in the fluid. Each sensors give unique optical signals under specific conditions that are acquired simultaneously by charge- coupled-device (CCD) optical detectors and those signal patterns are recognized as taste information. A novel micromachined structure has been added to the prototype structure for better performances in many parts. Surface micromachined structures confine sensor beads in the micromachined cavity as well as select designated sensor beads by way of size sorting method. This size selection method utilizes two separate size selective sieves on both sides of micromachined sensor cavities in the electronic taste chip. Each sensor is marked by certain size and selected by designated sensor cavity. Most preferred container of biochemical and chemical sensors is agarose beads. Agarose gel beads have the open pore structure which gives good attachment and binding to biological and chemical sensors. The structural characteristics of agarose bead are investigated and appropriate environment for agarose bead size selection method has been suggested. This research may be very useful in a micro total analysis system technology.Item Micromachining process for microsensor fabrication(Texas Tech University, 2001-08) Tang, XinMEMS (microelectromechanical systems) are an outgrowth of silicon technology. They are chip-sized systems that have the ability to affect their surroundings and to communicate what they know to the outside world in new and dramatic ways. And they can be made cheaply. In the fall of 1999, the Sensor Systems Center (SSC) was formed at Texas Tech University to capitalize on the growing interest in microsensors and microanalysis systems, the research was focused on the design, fabrication, and testing of wafer-based, compact microanalysis systems. At SSC, micromachining processes are under development for this purpose. Oxidation, lithography, PCD, CVD, etching are inherited from semiconductor processing and are done using processing tools and characterization equipments housed at the Maddox Laboratory. Bonding process is new and a tailored anodic bonding system has been self-built. Bonding between Si/Si, Si/Glass, and Glass/Glass wafer pairs can be done using these tools. Channels are made in silicon and glass substrate. A bulkmicomachinined accelerometer has been successfully fabricated. More complex and functional microsystems are being developed using these processes. In this paper these process steps are summarized with theoretical descriptions and actual fabrication examples, experimental results are investigated as well to optimize these process steps.Item Nanomaterials characterization and bio-chemical sensing using microfabricated devices(2004) Yu, Choongho; Shi, Li, Ph. D.A variety of nanostructured materials have been synthesized in recent years. These nanomaterials have potential applications in areas spanning computing, energy conversion, sensing, and biomedicine. Because of size confinement effects, furthermore, these nanomaterials are expected to show very different physical properties from those of their bulk counterparts. The measurement of their properties, however, has been very challenging due to their small dimensions. Similarly, it remains a challenge to detect chemical and biomolecular species due to their small dimensions. This dissertation presents the development of microelectromechanical systems (MEMS) devices for the characterization of thermophysical properties of nanomaterials and for the detection of chemical species and biological cells. The thermophysical property of one-dimensional (1D) nanomaterials was measured using a batch-fabricated microdevice consisting of two adjacent symmetric silicon nitride membranes suspended by long silicon nitride beams. Three methods were developed to assemble nanomaterials with the measurement devices. Those three methods include a wet deposition process, an in-situ chemical vapor deposition technique, and an electric-field-assisted assembly method. During the measurement, one membrane is Joule-heated to cause heat conduction through the nanomaterials to the other membrane, allowing for the measurement of thermal conductance and Seebeck coefficient. The electrical conductance can also be measured using the microdevice. The temperaturedependent properties of an individual single-wall carbon nanotubes (SWCNs) and SWCN bundles were measured. Measurement sensitivity, errors, and uncertainty were examined. The obtained thermal conductivity of an individual SWCN is found to be much higher than bundles of SWCNs in the range of 2000-11000 W/m-K at room temperature, in agreement with theoretical predictions. Furthermore, the thermal conductivity of bundles of SWCNs are found to be suppressed by contact resistance between interconnected SWCNs in the bundle. The microdevice has also been integrated with metal oxide nanobelts for chemical sensing. The sensing mechanism is based on surface oxidation-reduction (redox) processes that change the electrical conductance of the nanobelt. The sensor was found to be highly sensitive to inflammable and toxic gas species including nitrogen dioxide (NO2), ethanol, and dimethyl methylphosphonate (DMMP). Furthermore, it eliminated the sensor poisoning effects that have limited the wide use of polycrystalline metal-oxide based sensors. The experiment is a step towards the large scale integration of nanomaterials with microsystems, and such integration via an electric-field-directed assembly approach can potentially enable the fabrication of low-power, ultra-sensitive, and selective integrated nanosensor systems. The electric field manipulation technique has not only been used to assemble nanomaterials with MEMS, but also been used to focus biological cells in a microfluidic channel for cytometry applications. Flow cytometry is a powerful and versatile method of rapidly analyzing large populations of cells and other particulate or molecular analytes that have been captured on the surface of carrier particles. However, the key components of the system, hydrodynamic focusing and optical systems, make conventional cytometers complex, large, and expensive. To eliminate these drawbacks, a dielectrophoretic particle focusing technique combined with MEMS is explored to replace the hydrodynamic focusing mechanism. To focus particles, microelectrodes are patterned on the circumference of the channel to generate AC fringing fields that result in negative dielectrophoretic forces directing cells from all directions to the center of the channel. An ellipticlike microfluidic channel has been fabricated by isotropic etching of soda lime glass wafers and a subsequent wafer-bonding process. Experiments with microbeads and human leukemia HL60 cells and an analysis using a thin shell model indicate that biological cells can be focused using an AC voltage of an amplitude up to 15 Vp-p and a frequency below 100 kHz, respectively. This design eliminates the sheath flow and the fluid control system that makes conventional cytometers bulky, complicated, and difficult to operate, and offers the advantages of a portable standalone instrument as well as a module that could potentially be integrated with on-chip impedance or optical sensors into a micro total analysis system.Item Nonlinear control of an electrostatically actuated MEMS(Texas Tech University, 2003-12) Maithripala, SanjeevaOf the wide variety of actuation methods that have been developed for microelectromechanical systems (MEMS), electrostatic devices are the most common. The operation of these devices may be primarily categorized as digital and analog. Although the technology in digital operations is quite well developed the analog operations are not as developed due to an inherent nonlinear phenomenon in electrostatic actuation commonly known as "snap-through" "or pull-in." The concern of this thesis is analog control of electrostatically actuated MEMS devices. A simple 1-D model of electrostatic actuation is sufficient to capture the salient nonlinearities of the general problem. Several analog control schemes for such a model are proposed in the literature based on physical intuition. In the first part of the dissertation we show how these control schemes can be derived from the application of standard nonlinear control theory. Then we employ notions of feedback linearization, passivity and nonlinear state estimation to derive much improved control schemes that eliminates snap-through, improve performance with respect to low overshoot and faster settling times. The second part of the dissertation is concerned with generalizing the analog control notions developed for the 1-D model to a 3-D model where the device is assumed to freely rotate as well as translate. We first note that the system can be perceived as a coupled electromechanical system where the configuration space of the mechanical subsystem is the special Euclidian motion group SE{S). Thus we approach the problem from the setting of finding intrinsic control strategies for mechanical systems on general Lie groups. Furthermore, we do it in such a way that no coordinates need to be introduced on the Lie group. Thus, apart from the simplicity it provides, the tools developed in here may be of great importance to long term trajectory planning and optimal control problem on Lie groups and stabilization and active vibration absorption of rigid 3-D structures.Item Optical deformability : micromechanics from cell research to biomedicine(2001-12) Guck, Jochen Reinhold; Käs, Josef A.Item Passivity-based stabilization of a 1-DOF electrostatic MEMS model with a parasitic capacitance(2008-05) Wickramasinghe, Imiya M.; Berg, Jordan M.This thesis addresses the problem of stabilizing 1-DOF piston mode electrostatic actuator in the presence of parasitic capacitance due to conductive substrate. The current study makes use of passivity-based control technique to formulate controllers. The static and dynamic controller schemes based on total charge can result in a unique equilibrium, however, their region of attraction may be small and the equilibrium may lose stability through Hopf bifurcation for certain configurations. A new charge quantity Qcc is introduced and used to derive static and dynamic feedback controllers in order to resolve issues encountered in the controllers based on total charge. The controllers based on Qcc are proved to be capable of globally asymptotically stabilizing the unique feasible equilibrium point for the configurations where the movable electrode is screened from the parasitic electrode by the control electrode. When the movable electrode and the parasitic electrode are directly coupled to have a mutual capacitance, numerical simulations show that the region of attraction of the closed-loop equilibrium is large. The effects of the infinite parallel plate approximation inherent in the formulation on the controller performance are investigated.Item Probabilistic modeling of microelectromechanical systems (MEMS)(Texas Tech University, 2002-12) Khandaker, Morshed P. H.Micro-Electro-Mechanical Systems (MEMS) are a fast-developing technology that have a potential to permeate most engineering and medical applications. For this technology to continue expanding, issues regarding the cost of manufacturing and reliability of the devices have to be addressed. To improve the reliability, probabilistic design methodologies are potent in both the modeling and testing of high-performance MEMS. The benefit of probabilistic design approaches is a more rational basis for making design decisions that balance component or system efficiency with reliability or safety. Probabilistic methods are used to assess uncertainties involved in the manufacturing of MEMS devices. Probabilistic methods guide the design of these devices to achieve reliable design in a most efficient way. The objectives of the research work were to formulate and analyze probabilistic failure criteria on a simplified capacitive accelerometer mode. In this respect, comprehensive probabilistic and deterministic analysis was carried out for the selected model. The scope of work is threefold. First, two probabilistic failure criteria will be investigated on the capacitive structure, namely probabilistic clearance failure criterion and probabilistic fracture toughness failure criterion. Second, four kinds of probabilistic analyses for characterization of MEMS will be used: probability of failure, sensitivity analysis, safety index, and probability-based design. Third, three kinds of finite element analyses, namely static, modal and spectral analysis, will be used to see the deterministic response and will be compared with probabilistic resultItem Smart microplates: integration of photodiode within micromachined silicon pyramidal cavity for detecting chemiluminescent reactions and methodology for passive RFID-type readout(2007-12) Park, Yoon Sok, 1977-; Neikirk, Dean P., 1957-Since the late 1990s our group has been working with groups in chemistry department at the University of Texas at Austin on a project referred as "Electronic Taste Chip," a MicroElectroMechanical System (MEMS) based miniaturized microfluidic chemical sensor with multianalyte detection capabilities. By integrating optical detection mechanism directly onto the silicon chip a cost effective, compact, and portable sensor can be realized enabling use of these chips out of conventional laboratory environment. Addition to the integration a noble approach of accessing a photodiode with non-contact powerless RFID type readout is presented. By doing so a packaged photodiode can be interrogated without direct electrical contact, enhancing the portability even further for a sensor operated in aqueous medium. First background information regarding the project as well as design and integration criteria is presented followed by demonstration of non-contact RFID-type readout of a photodiode. Detailed discussion on the development of process integration scheme is discussed along with the measurements verifying the performance of the fabricated photodiode. During this investigation normally overlooked design criteria of collection efficiency, the effect of how a target element is to be delivered to a detection mechanism on the overall performance of the sensor, is addressed and discussed.Item Test and characterization of engineering nanocoatings for mems and nanoenergetic materials(2012-05) Vijayasai, Ashwin; Dallas, Timothy E. J.; Gale, Richard O.; Pantoya, Michelle; Weeks, Brandon L.; Hase, William L.; Yeo, ChangdongThis dissertation presents the development, test and characterization of engineered nanocoatings for MEMS and nano-energetic reactive materials. Surface modification on MEMS and nano-energetic reactive materials are developed using a commercially available nanocoating tool. Surface modifications include Chemical Vapor Deposition of Fluorocarbon SAM and nanoparticles and Atomic Layer Deposition of thin oxides. Detailed descriptions of the nanocoating process and their chemical reactions are explained. An F-SAM coated MEMS tribogauge is characterized to estimate the adhesive and frictional forces. In-situ frictional measurements were made. Increasing adhesion force was observed for increasing number of load cycles. The tribogauge is later used as an ex-situ characterization tool to observe the performance of various nanocoating recipes for F-SAM coating. Characterization of the tribogauge is performed using an electronic sense tool. Contact angle goniometer was used to characterize the performance of various recipes. Various types of nanocoatings were deposited on witness samples and nano-energetic materials. A comparison study of underwater combustion tests were made on these thermite pellets. An aging study was performed on both nanocoated witness samples and pellets. The aging experiment is performed by submerging them in de-ionized water for 10 days. Contact angle goniometer and few optical microscopes were used to characterize the performance of various recipes. Apart from the nanocoating based projects, this dissertation briefly explains other projects that were part of the graduate program. A brief description and initial results of a few MEMS device designs are explained in this dissertation. As part of future work new MEMS devices were designed that will allow follow-up nanocoatings projects.Item Testing and characterization of 360° rotating micromirrors(2009-08) Oak, Sahil; Dallas, Timothy E. J.; Gale, Richard O.The out-of-plane micromirrors have proved to be a vital component of Micro-Opto-Electro-Mechanical Systems (MOEMS).These systems have been developed for wide range of applications including optical switching, beam steering and precise transmission and reception of bio-optical signals. This thesis focuses on design, test and characterization of rotating out-of-plane micromirrors. The system consists of polysilicon micro-mirror which is held at 45° angle on gear which has a rotational freedom of 360°. The mirror assembly is pushed into position using an assembling probe. Actuation of the rotating structure is accomplished via a coupled torsional ratcheting actuator (TRA). A test setup was developed for the characterization of the mirror in which the average intensity of reflected light is measured. Also, effect of TRA ratcheting events on the gear rotation is studied. A design for auto pop-up micro-mirror is discussed which allows the system to self assemble by using force from electrostatic actuator (TRA). However, the mechanism was not successfully demonstrated. Failure analysis of this design is done and alternative design with improvements is suggested.Item The impact packaged device planarity has on insertion loss in a DMD based optical networking system(Texas Tech University, 2003-12) Spain, James DThis paper will effectively show whether or not packaged die planarity has an impact on optical system insertion loss and also determine whether or not the die attach systems compromises system performance criteria due to reduced packaged die planarity. A new and important application for Texas Instruments Incorporated Digital Micromirror Device (DMD^M) is in optical switching where it can play a role in dynamic network reconfiguration and signal grooming in the DWDM technology. Using the DMD™ will help the Dense Wavelength Division Multiplexing (DWDM) technology reach a high number of wavelengths in the optical fiber, thus increasing the overall bandwidth of the optical fiber. To obtain this goal, many parameters must be considered. One key performance parameter in this application is insertion loss. Insertion loss is the light power that is lost going from one optical fiber to another as it passes through key optical components. In this work, the impact of device planarity on insertion loss for a particular system configuration is investigated. Calculating the radius of curvature from the topography of the packaged die will enable the user to determine the mathematical approximation of the impact of planarity on insertion loss. Interferometric data obtained by measuring the packaged devices are used to create a simple model for the optical performance. The insertion loss of the DMDTM in a generic fiber optic system can be estimated. The distributions of two different die attach systems are compared from an optical performance perspective.