Experimental Noise Analysis of Reed Switch Sensor Signal under Environmental Vibration

Similar documents
Model Display digit Size Output Power supply 24VAC 50/60Hz, 24-48VDC 9999 (4-digit) 1-stage setting

PROJECT #2 GENERIC ROBOT SIMULATOR

Indicator No mark Single preset Dual preset DIN W144 H72mm DIN W48 H96mm No mark DIN W72 H72mm (4 digit) (6 digit) Counter/Timer

Design of FPGA- Based SPWM Single Phase Full-Bridge Inverter

Measurement of Equivalent Input Distortion AN 20

GFM. Typical Aluminum GFM Mass Flow Meter NIST MASS FLOW METERS. Principles of Operation. Design Features. n n n n n n n n n n

E X P E R I M E N T 13

INCREASE OF STRAIN GAGE OUTPUT VOLTAGE SIGNALS ACCURACY USING VIRTUAL INSTRUMENT WITH HARMONIC EXCITATION

Compound Controller for DC Motor Servo System Based on Inner-Loop Extended State Observer

Design of FPGA Based SPWM Single Phase Inverter

By: Pinank Shah. Date : 03/22/2006

MEASUREMENT AND CONTORL OF TOTAL HARMONIC DISTORTION IN FREQUENCY RANGE 0,02-10KHZ.

GENERATE AND MEASURE STANDING SOUND WAVES IN KUNDT S TUBE.

Single Bit DACs in a Nutshell. Part I DAC Basics

APPLICATION NOTE UNDERSTANDING EFFECTIVE BITS

DIGITALLY TUNED SINUSOIDAL OSCILLATOR USING MULTIPLE- OUTPUT CURRENT OPERATIONAL AMPLIFIER FOR APPLICATIONS IN HIGH STABLE ACOUSTICAL GENERATORS

Electronic motor protection relay

TMCM BLDC MODULE. Reference and Programming Manual

Tehrani N Journal of Scientific and Engineering Research, 2018, 5(7):1-7

AC : USING ELLIPTIC INTEGRALS AND FUNCTIONS TO STUDY LARGE-AMPLITUDE OSCILLATIONS OF A PENDULUM

AkinwaJe, A.T., IbharaJu, F.T. and Arogundade, 0.1'. Department of Computer Sciences University of Agriculture, Abeokuta, Nigeria

Data Acquisition System for Electric Vehicle s Driving Motor Test Bench Based on VC++ *

Laboratory Exercise 3: Dynamic System Response Laboratory Handout AME 250: Fundamentals of Measurements and Data Analysis

Technical Explanation for Counters

Design and Construction of a Three-phase Digital Energy Meter

A New Design of Log-Periodic Dipole Array (LPDA) Antenna

A New Basic Unit for Cascaded Multilevel Inverters with the Capability of Reducing the Number of Switches

Outline. Motivation. Analog Functional Testing in Mixed-Signal Systems. Motivation and Background. Built-In Self-Test Architecture

Analysis of SDR GNSS Using MATLAB

A Novel Small Signal Power Line Quality Measurement System

A SELECTIVE POINTER FORWARDING STRATEGY FOR LOCATION TRACKING IN PERSONAL COMMUNICATION SYSTEMS

Objectives. Some Basic Terms. Analog and Digital Signals. Analog-to-digital conversion. Parameters of ADC process: Related terms

SELECTION AND CONNECTION OF SPRING APPLIED FAILSAFE AND PERMENANT MAGNET BRAKES

COMPRESSION OF TRANSMULTIPLEXED ACOUSTIC SIGNALS

RISH Clamp POWER 1000A / 400A AC-DC. Application. Product Features. Unique Design. Non Contact Voltage Detection. Three Phase Power Measurement

High-Order CCII-Based Mixed-Mode Universal Filter

Application of Improved Genetic Algorithm to Two-side Assembly Line Balancing

Radar emitter recognition method based on AdaBoost and decision tree Tang Xiaojing1, a, Chen Weigao1 and Zhu Weigang1 1

CHAPTER 5 A NEAR-LOSSLESS RUN-LENGTH CODER

Fault Diagnosis in Rolling Element Bearing Using Filtered Vibration and Acoustic Signal

INF 5460 Electronic noise Estimates and countermeasures. Lecture 11 (Mot 8) Sensors Practical examples

Problem of calculating time delay between pulse arrivals

HB860H 2-phase Hybrid Servo Drive

KMXP SERIES Anisotropic Magneto-Resistive (AMR) Linear Position Sensors

doi: info:doi/ /ifeec

Fingerprint Classification Based on Directional Image Constructed Using Wavelet Transform Domains

AME28461 SERIES EMI FILTER HYBRID-HIGH RELIABILITY

A study on traffic accident measures in municipal roads by using GIS

Metrology and Quality Control

AME50461 SERIES EMI FILTER HYBRID-HIGH RELIABILITY

Delta- Sigma Modulator with Signal Dependant Feedback Gain

Sensors & Transducers 2015 by IFSA Publishing, S. L.

hi-rel and space product screening MicroWave Technology

Vilnius Gediminas Technical University Naugarduko 41, LT-03227, Vilnius, Lithuania

Multisensor transducer based on a parallel fiber optic digital-to-analog converter

Smart Energy & Power Quality Solutions. ProData datalogger. Datalogger and Gateway

R. W. Erickson. Department of Electrical, Computer, and Energy Engineering University of Colorado, Boulder

SERCOS ENERGY. SERCOS International e.v.

A New Space-Repetition Code Based on One Bit Feedback Compared to Alamouti Space-Time Code

Selection of the basic parameters of the lens for the optic-electronic target recognition system

Indoor Localization by Pattern Matching Method Based On Extended Database

Inertial Technology For North Finding Steve Clarke, Applications Engineer.

Calibrating Car-following Model with Trajectory Data by Cell Phone

RCP2W RC AW RCS2W. Dust- proof/splash-proof Type. 441 Dust-proof/Splash-proof. Air-Oil Systems, Inc. Dust-proof/Splash-proof Type

Measurements of weld geometry using image processing technology

HOW BAD RECEIVER COORDINATES CAN AFFECT GPS TIMING

NOISE IN A SPECTRUM ANALYZER. Carlo F.M. Carobbi and Fabio Ferrini Department of Information Engineering University of Florence, Italy

RISH CON - Hz FREQUENCY TRANSDUCER

Chapter 1 The Design of Passive Intermodulation Test System Applied in LTE 2600

LIQUIDS AND GAS FLOW REGULATORS. Design Features. n n n n. Regulator Systems. Normally closed (NC) when de-energized.

PRACTICAL ANALOG DESIGN TECHNIQUES

CAEN Tools for Discovery

The Firing Dispersion of Bullet Test Sample Analysis

Intelligent IMC-PID Control Using PSO for Ultrasonic Motor

High Speed Area Efficient Modulo 2 1

202 Chapter 9 n Go Bot. Hint

HELIARC. THE FIRST NAME IN TIG.

New MEGA POWER DUAL IGBT Module with Advanced 1200V CSTBT Chip

PRACTICAL FILTER DESIGN & IMPLEMENTATION LAB

MULTIMETER FOR POWER AND ENERGY MEASUREMENT HIGH-END DIGITAL MULTIMETER

Measurements of the Communications Environment in Medium Voltage Power Distribution Lines for Wide-Band Power Line Communications

A New 3-Bit Integrating Time to Digital Converter Using Time to Voltage Conversion Technique

Massachusetts Institute of Technology Dept. of Electrical Engineering and Computer Science Fall Semester, Introduction to EECS 2.

CCD Image Processing: Issues & Solutions

Motor Vibration Detect Mechanical & Electrical Motor Faults with Vibration Monitoring Instrumentation

The Detection of Abrupt Changes in Fatigue Data by Using Cumulative Sum (CUSUM) Method

Super J-MOS Low Power Loss Superjunction MOSFETs

Previous R&D of vibrating wire alignment technique for HEPS

Min. Meas. Range flow velocity = 0 to 1.64 ft/s

High Temperature Accelerometers

Optimal Arrangement of Buoys Observable by Means of Radar

Intermediate Information Structures

Analysis and Optimization Design of Snubber Cricuit for Isolated DC-DC Converters in DC Power Grid

Simulation and Analysis on Signal Acquisition of BDS Receiver with The Aid of INS

Estimation of reflection location by the correlation coefficient function

Subscriber Pulse Metering (SPM) Detection

Zonerich AB-T88. MINI Thermal Printer COMMAND SPECIFICATION. Zonerich Computer Equipments Co.,Ltd MANUAL REVISION EN 1.

Electro-Magnetic Flowmeters MINI-MAG W Vibration Limits: MAGNETIC FLOWMETER (with Integral XE Converter) Power Requirements: Power Consumption:

A GHz Constant KVCO Low Phase Noise LC-VCO and an Optimized Automatic Frequency Calibrator Applied in PLL Frequency Synthesizer

4. INTERSYMBOL INTERFERENCE

Transcription:

Computer Techology ad Applicatio 7 (16) 96-1 doi: 1.1765/1934-733/16..4 D DAVID PUBLISHING Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio Odgerel Ayurzaa 1 ad Hiesik Kim 1. Departmet of Computer Egieerig, Mogolia Uiversity of Sciece ad Techology, Ulaabaatar, Mogolia. Departmet of Electrical ad Computer Egieerig, Uiversity of Seoul, Seoul, Korea Abstract: The chatterig oise problem of reed switch sesor sigal for Automatic Meter Readig system was aalyzed experimetally uder various types of exteral vibratios ad shocks. The exteral vibratio level amplitude was measured with a accelerometer. To apply for water flow measuremet devices, the reed switch sesors should keep high reliability. But the measured digital meter data are occurred differece or errors by chatterig oise. The reed switch cotais chatterig error by itself at the force equivalet positio. The vibratios such as passig vehicle ear to the reed switch istalled locatio causes chatterig. I order to reduce chatterig error, most system uses just software methods, for example usig digital filter algorithm ad also statistical calibratio methods. However software approaches were implemeted for reducig chatterig error, there has still geerated chatterig error due to exteral mechaical vibratios ad magetic field. The chatterig errors ca be reduced by chagig leaf sprig structure usig mechaical hysteresis characteristics. Key words: Noise aalysis, accelerometer, maget, vibratio, reed switch, sesor, chatterig error. 1. Itroductio To measure water flow amout for AMR (automatic meter readig) system of water supply, a reed switch, CYBLE, MR (magetic resistive) ad CCD (charge couple device) image sesors are applied i field. The covetioal water flow meters of mechaical type have implied may limitatios that a perso eeds to visit the locatio of the each water meter ad write dow maually the meter data [1]. Maual readig takes much time, labor ad expeses. Each of this method has its ow advatages ad disadvatages. The MR, CYBLE ad image sesor modules are cosumed the additioal battery. The reed switch sesor rus o without battery that is directly coected to the termial device. The reed switch sesor has very simple iteral ad exteral structure ad is easy to apply ad mout for ay practical fields. Correspodig author: Odgerel Ayurzaa, Ph.D., research fields: sesor etworkig, embedded systems ad applicatios, sigal processig. Hiesik Kim, Ph.D., research fields: optical measuremet of geometries, applicatios of sesors for automatio, image processig. Also reed switch sesor is much cheaper tha other sesor modules. The reed switch sesor cotais chatterig error seriously by itself iteral structure. The purpose of this study is focused o aalysis ad reductio of chatterig error for remote measuremet of water supply maagemet. The operatio of reed switch sesors ca be described as a mechaical cotact switch by approximatio of permaet maget piece to geerate a electrical pulse. The reed switch sesors are used mostly i measuremet applicatio to detect the rotatioal or traslatioal displacemet. They are applied for the electroic digital type of water flow meters. The reed switch sesor is just mouted simply o the covetioal mechaical type flow meter. A small maget is attached o a poiter of the water meter couter rotor. Iside the reed sesor, two steel leaf sprigs create mechaical cotact ad apart repeatedly as rotatio of flow meter couter. The coutig electrical cotact pulses ca be coverted as the water flow amout. The MCU seds the digital flow rate data to the server usig the wireless commuicatio etwork. But the digital meter data are

Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio 97 occurred differece or errors by chatterig oise. Two ickel iro blades, sealed hermetically i a glass tube are formig together a cotact uit, hereby the magetically part is equivalet to the electrical part, wheever a magetic field, costructed by a coil or a permaet maget is applied to the termials, magetic orth ad south poles are coducted to the cotact zoe, with the effect that the blades are pulled together. Wheever the magetic field is weak eough, the blades are released agai. The differece betwee pull i ad drop out is amed the hysteresis ad is determied by the thickess of the o-magetic surface of the cotact zoe ad of the overlap of these blades. With this behavior the reed cotact has a desigated sap characteristic. I Ref. [] paper presets cotact materials for reed switches. I Ref. [3] paper discusses the miiaturizatio limit of the reed switch with respect to the magetic ad mechaical properties. Ifluece of cotact surface coditios of reed switches is ivestigated i Ref. [4]. To reduce chatterig i slidig-mode cotrol, a boudary layer aroud the switchig surface is used, ad a cotiuous cotrol is applied withi the boudary [5]. A covetioal reed switch is operated by the mechaical force, so there has cotaied a desig limitatio. There is geerated error by the actual state vibratio whe magetic ad reed switch are approximated ad separated due to force balace coditio. Due to the magetic attached poiter of the water meter couter rotor stops ear to the reed switch ad also exteral impacts ad vibratios such as passig vehicle ear to the reed switch istalled locatio causes additioal oise pulses that amed as chatterig error. I order to remove these additioal oise pulses reset systems use just software methods, for example usig digital filter algorithm ad statistical calibratio methods. But there is ecessary eed to itroduce hardware approaches that should upgrade itself hardware structure for reed switch. Some small particles are blocked iside to the gear ad wig of the mechaical water flow meter whe water flows through the water meter the rotatig wig is possible to jam ad break dow.. Backgroud of Study Seoul water works research istitute supported the project amed as Itroductio strategy for automatic meter readig system for waterworks services durig oe year. As a result of this project, the meter data receive rate usig reed switch sesor module which was lower tha other sesor modules. As show i the experimet, there was occurred water data differece betwee remote ad huma power measuremet whe implemetig the reed switch sesor module for water automatic measuremet system. Water supply compaies have developed ad implemeted their ow techology for remote measuremet system. As followig Fig. 1 shows average meter data receive rates of various kids of sesor modules which were experimeted for remote measuremet of water supply. Above metioed project result of oe year, the average meter data receive rate usig reed switch sesor module was lower tha digital MR ad CYBLE sesor modules. The meter data receive rate usig the reed switch sesor was decreased due to the chatterig error that causes various types of vibratios ad magetic fields [6]. Therefore, there eeds a study to aalyze ad reduce Fig. 1 Compariso of meter data receive rates.

98 Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio the chatterig error. The chatterig meas the udesired itermittet whe a momet of makig ad breakig operatio of reed sprigs. It may occur either whe the reed cotact is operated, released or whe subjected to exteral shock or vibratio. 3. System Desig for Chatterig Error Followig a sub-miiature (KSM-13) ad a low-hysteresis (KMC-185) two types of reed switches were used for chatterig experimet. KSM-13 is a sub-miiature reed switch. This forms a reed switch which is built with a.5 mm diameter glass havig a higher wall thickess for better glass to metal seals, ad is a good solutio for geeral purpose, high volume applicatios with a little shock ad vibratio. The ceter cotact also makes it suitable for coil applicatios. The three differetial rages available cover a wide rage of releasig specific applicatios. KMC-185 is a low-hysteresis reed switch that is built with specially pressed sprigs with slightly higher rigidly for close differetial, low hysteresis applicatios where a operate ad release are required with miimum maget travel or miimum chage i coil voltage. I order to reduce the hysteresis characteristics, two reed sprigs of the reed switch are mouted perpedicular each other that meas the cotact area of the two reed sprigs less tha other types of reed switches. Fig. 3 shows the mai schematic diagram of the vibratio experimet system. A air compressor (SEN 5-1, max pressure 8 kg/cm ) cotrolled vibrator was desiged as a rigid plastic platform. The vibrator system cosists of air compressor, air pressure regulator (5 kg/cm ), pressure meter, soleoid valves, air cylider tube, left ad right velocity regulator valves, left ad right limit switches, rail guide ad movig table. The air compressor has bee provided by high pressure, i our case 5 kg/cm that ca be regulated air pressure amout by the pressure regulator (PW ). Two coectors of the soleoid valve (TVF 33) were coected with air cylider (TCP 1B16-45) by hose (Ø 6mm). The iside of the air cylider has movig steel shaft Fig. Iteral structure of reed switches (KSM-13 ad KMC-185). Fig. 3 Experimetal schematic layout of mai oscillatio table of reed switch vibratio experimet.

Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio 99 (45 mm) for makig lie motio. This steel shaft of the air cylider is coected to the movig oscillatio table directly for creatig the vibratio. The vibratio velocity has bee cotrolled by left ad right regulator valves that are mouted o the air cylider. That meas to regulate high ad low air pressure from air compressor. Left ad right limit switches are istalled at the left ad right eds of movig table respectively. The movig table had a size of 181 mm makig by acrylic plastic. The acceleratio sesor, reed switch sesor ad small maget of ferrite type (115 mm, magetic flux desity B = 633.6 Gauss) were istalled o the movig table for measurig vibratio level ad chatterig error. I order to measure the vibratio level, system has utilized a 3 axes acceleratio sesor (MMA76Q). A surface-mouted accelerometer was attached to the plastic acrylic platform to measure horizotal ad vertical acceleratio of the movig table. The accelerometer is a DC accelerometer havig a rage of 6 g. g is the acceleratio of gravity equals to 9.8 m/sec. The ATMega 18 kit was used to covert the reed switch, vertical ad horizotal vibratio aalog sigals to the digital sigals through the ADC (F port). The iput clock frequecy of the ADC is 15 KHz. The RS3 port rate, data bit, stop bit ad parity were set 57,6 bps, 8 bit, 1 bit ad o parity respectively. As show i the experimet, 358 samplig umbers of chatterig error data were icluded 3 sec. I our case, we have estimated couts of chatterig error pulse per every 1 secods. After removig oise of the acceleratio sesor output, it is ecessary to estimate proper vibratio amplitude. Vibratio amplitude values have to cover all cotets ad characteristic of the vibratio. The RMS approach was used to estimate for quatifyig the vibratio level. The RMS for a collectio of values x, x,... x } is { 1 The correspodig formula for a cotiuous fuctio f(t) defied over the iterval T1 t T is f RMS T 1 T T 1 T 1 [ f ( t) ] dt () The RMS of a periodic fuctio is equal to the RMS of oe period of the fuctio. The RMS value of a cotiuous fuctio or sigal ca be approximated by takig the RMS value of a series of equally spaced samples. 4. Experimet ad Aalysis of Chatterig Error The output sigal mixed with much oise whe a RC low pass filter is ot coected o the outputs of the acceleratio sesor X, Y, Z. I order to remove white radom oises, RC low pass filter was implemeted i the circuit of accelerometer. Accordig to the iput vibratio level, there is geerated the additioal oise pulses amed as chatterig error. There were seriously geerated the chatterig error pulses whe the acceleratio value icreases. Our experimet device usig the air compressor has created much strog vibratio levels. As show i the experimet, maximum ad miimum vibratio levels were.18 g ad.11 g by RMS method respectively. We caot measure low level of vibratio util.11 g. Therefore, i order to measure low level of vibratio x RMS 1 x1 x... x xi i1 (1) Fig. 4 Low level vibratio ad chatterig error.

Amplitude (Volt) Amplitude (Volt) Amplitude (Volt) Amplitude (Volt) 1 Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio 1.6 Acceleratio Sesor X Value 1.5 1.4 1.3 5 1 15 5 Samplig umber Reed S/W Sesor Chatterig Error 5 4 3 Fig. 5 High level vibratio ad chatterig error. 1 5 1 15 5 Samplig umber Fig. 8 Geeratig chatterig error whe vibratio level was.15 g (KSM-13). Fig. 6 Chatterig experimet table usig rubber cords. 1.5 Acceleratio Sesor X Value 1.5 1.48 1.46 5 1 15 5 Samplig umber Reed S/W Sesor Chatterig Error 5 4 3 1 5 1 15 5 Samplig umber Fig. 7 No chatterig error whe vibratio level was.11 g (KSM-13). there is eed to desig aother motio force. The rubber cords were fixed right ad left eds of the movig oscillatio table for creatig low level vibratio like followig Fig. 6. A perso makes vibratio with her/his had mechaically. I further experimet, a low level of vibratio util.11 g was measured usig oscillatio table as show i Fig. 6. The vibratio table usig the air compressor is used from.11 g vibratio level value. Fig. 9 Layout of a 3-axis accelerometer ad a reed switch i X directio (horizotal vibratio). It s very importat to describe begiig vibratio level value of chatterig error by precisely. The chatterig error pulses were ot geerated whe vibratio level value icreases util.11 g for KSM-13. The chatterig error pulses were geerated from vibratio level value of.15 g. 5. Result for Chatterig Error The cotactig distace betwee reed switches (KSM-13, KMC-185) ad ferrite magets (B = 633.6 Gauss) was fixed 11 ad 7 mm respectively. This distace of 11 ad 7 mm is a cotact momet (O ad Off) of the reed sprigs whe the maget approaches the sesor. As show i Fig. 1, vibratio bear capability of the low-hysteresis reed switch was better tha sub-miiature reed switch.

Number of chatterig error pulses Number of chatterig error pulses Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio 11 1 1 8 6 Chatterig error for horizotal vibratio ad Y directio begiig RMS values of vibratio for low-hysteresis reed switch (KMC-185) respectively. As show i the graphics, low-hysteresis reed switch was more stable agaist the exteral vibratios tha sub-miiature reed switch. 6. Coclusios 4 Sub-miiature Reed SW (KSM-13) Low-hysteresis Reed SW (KMC-185)..4.6.8 1 1. 1.4 1.6 Fig. 1 Chatterig error of reed switch for horizotal vibratio. Fig. 11 Layout of a 3-axis accelerometer ad a reed switch i Y directio (vertical vibratio). 1 1 8 6 4 Acceleratio sesor RMS value [g=9.8 m/s ] Chatterig error for vertical vibratio Sub-miiature Reed SW (KSM-13) Low-hysteresis Reed SW (KMC-185)..4.6.8 1 1. 1.4 1.6 Acceleratio sesor RMS value [g=9.8 m/s ] Fig. 1 Chatterig error of reed switch for vertical vibratio. As show i Figs. 1 ad 1, X directio vibratio is more sesitive (geeratig more chatterig error pulses) tha Y directio vibratio because vibratio directio axis is perpedicular to the reed sprigs of the reed switch. Ad.1117 g ad.158 g are the X The chatterig error of reed switch sesor for AMR system was aalyzed accordig to various types of exteral vibratios ad shocks. Two kids of reed switches were aalyzed for vibratio experimet. Low hysteresis reed switch was more stable tha stadard type reed switch agaist vibratios. As show i a experimet result, X axis directio vibratio is more sesitive (geeratig more chatterig error pulses) tha Y axis directio vibratio. I further work, we have to desig better performig reed switch agaist the vibratios such as desigig or 3 reed switches mouted o the mechaical type water flow meter. Also other approach is to chage iside sprig leaf structure of reed switch. Ackowledgmet This research was fiacially supported by the Project No. 15163 Sesitive ad Errorless Electrical Security Fece by Usig Miimal Frictio Electricity i Coaxial Cable of the Natioal Research Foudatio of Korea ad by the Project No. COR_1/15 High Sesitive ad Errorless Security Fece System of the Mogolia Foudatio for Sciece ad Techology ad by the Project No. 13414 Security Fece Detectio Accuracy of Cable Frictio Electricity Sesor System of the Uiversity of Seoul. Refereces [1] Ayurzaa, O., Kim, H., Park, Y., ad Kwo, J. 7. Ubiquitous Telemetrig System for Water Flow Meters Usig Zigbee Module ad CDMA Modem. ICUCT, Jauary, 8-31, Aomori, Japa. [] Jah, W., ad Guther, W. 1995. Cotact Materials for Reed Switches. [3] Kobayashi, T., Hiohara, K., ad Kawakita, C. 1991.

1 Experimetal Noise Aalysis of Reed Switch Sesor Sigal uder Evirometal Vibratio Magetic Mechaical Desig of Ultramiiature Reed Switches. IEEE 1991. [4] Hiohara, K., Kobayashi, T., ad Kawakita, C. 1993. Ifluece of Cotact Surface Coditios o Discharge Pheomea of Vacuum Reed Switches. IEEE Tras. Compoet, Hybrids ad Maufacturig Techology 16 (). [5] Kachroo, P., ad Tomizuka, M. 1996. Chatterig Reductio ad Error Covergece i the Slidig-Mode Cotrol of a Class of Noliear Systems. IEEE Tras. o Automatic Cotrol 41 (7).