1、INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY Contamination Control Division Reco ended Practice 024.1 IEST-RP-C024.1 Measuring and Reporting Vibration in icroelectronics Facilities INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY Arlington Place One 2340 S. Arlington Heights Road, Suite 10 A
2、rlington Heights, IL 6005-4516 Phone: (847) 981-010 Fax: (847) 981-4130 E-mail: iestiest.org Web: ww.iest.orgThis Recomended Practice is published by the INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY to advance the technical and enginering sciences. Its use is entirely voluntary, and determinat
3、ion of its aplicability and suitability for any particular use is solely the responsibility of the user. This Recomended Practice was prepared by and is under the jurisdiction of Working Group 024 of the IEST Contamination Control Division. Copyright 194 by the INSTITUTE OF ENVIRONMENTAL SCIENCES AN
4、D TECHNOLOGY Fifth printing, December 202 ISBN 978-1-87862-24-3 PROPOSAL FOR IMPROVEMENT: The Working Groups of the INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY are continualy working on improvements to their Recomended Practices and Reference Documents. Sug-gestions from those who use these d
5、ocuments are welcome. If you have a sugestion regarding this document, please use the online Proposal for Improvement form found on the IEST website at ww.iest.org/proposal/form.html. INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY Arlington Place One 2340 S. Arlington Heights Road, Suite 10 Arli
6、ngton Heights, IL 6005-4516 Phone: (847) 981-010 Fax: (847) 981-4130 E-mail: iestiest.org Web: ww.iest.org IEST-RP-CC024.1 INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY 3CopyrightedmaterialMeasuring and Reporting Vibration inMicroelectronics FacilitiesIEST-RP-CC024.1CONTENTSSECTION1 SCOPE AND L
7、IMITATIONS 52 REFERENCES . 53 TERMS AND DEFINITIONS 64 BACKGROUND: OVERVIEW OF THE PROCESS OF ESTABLISHINGVIBRATION CRITERIA 85 INSTRUMENTATION FOR MEASURING VIBRATION . 96 PROCEDURES FOR MEASURING VIBRATION 11TABLE1Vibration measurement and data set-up checklist. 15IEST-RP-CC024.1 INSTITUTE OF ENVI
8、RONMENTAL SCIENCES AND TECHNOLOGY4CopyrightedmaterialIEST-RP-CC024.1 INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY 5CopyrightedmaterialINSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGYContamination Control DivisionRecommended Practice 024.11.2 LimitationsThis Recommended Practice focuses on vi
9、bration.Although control of acoustical disturbances isequally important, it is not the subject of this RP.However, ANSI documents S1.1, S1.4, S1.11, andS1.13 (see section 2) are available for those seekingguidance in measuring acoustical noise.2 REFERENCESANSI-S1.1: Acoustical Terminology, Including
10、 Mechan-ical Shock and Vibration. American National Stan-dards Institute.ANSI-S1.4: Specification for Sound Level Meters. Amer-ican National Standards Institute.ANSI-S1.11: Specification for Octave-Band and Frac-tional-Band Analog and Digital Filters. AmericanNational Standards Institute.ANSI-S1.13:
11、 Methods for the Measurement of SoundPressure Levels. American National StandardsInstitute.Bendat, J.S., and A.G. Piersol: Random Data: Analy-sis and Measurement Procedures. 2nd ed., Wiley-Interscience, New York: 1986.Biggs, J.M.: Introduction to Structural Dynamics.McGraw-Hill, New York: 1964.Ewins
12、, D.J.: Modal Testing: Theory and Practice. Re-search Studies Press, Letchworth, England: 1984.1 SCOPE AND LIMITATIONS1.1 ScopeEquipment used in the manufacture, measurement,and inspection of integrated circuits is sensitive tovibration and sound. It is therefore necessary toestablish levels of vibr
13、ation sensitivity for themand to ensure that vibrations occurring in the facil-ity or at the site at which they are located are belowthose levels.Such equipment will be referred to as process toolsin this Recommended Practice (RP).This RP is intended to provide guidance for use inthe microelectronic
14、s industry. However, it mayalso be applicable in pharmaceutical and biologicalresearch, metrology laboratories, and other con-texts in which vibration control is important.This RP provides guidelines for conducting vibra-tion measurements and reporting vibration data,specifically for:a) Selecting su
15、itable instrumentation and hard-ware for use in vibration measurements.b) Establishing sensitive process tool vibrationthresholds.c) Conducting vibration measurements on sitesand in facilities.d) Reporting results of vibration measurements ina uniform and consistent format.Measuring and Reporting Vi
16、bration inMicroelectronics FacilitiesIEST-RP-CC024.1IEST-RP-CC024.1 INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY6CopyrightedmaterialHurty, W.C., and M.F. Rubenstein: Dynamics ofStructures. Prentice-Hall, Englewood Cliffs, NewJersey: 1964.Oppenheim, A.V., and A.S. Willsky: Signals andSystems. P
17、rentice-Hall Signal Processing Series,Englewood Cliffs, New Jersey: 1983.3 TERMS AND DEFINITIONSMany of the definitions in this list are specialized tothis document.aliasingA distortion of frequency data that occurs duringdigital data acquisition when the sampling rate ofthe data acquisition hardwar
18、e is too slow to capturethe higher-frequency data that may be present. (Ananti-aliasing filter to block out high frequency datamay be used to overcome this phenomenon.)averagingThe process of developing the mean average valueof a signal.centile averagingThe process of determining the statistical tim
19、edistribution of an ensemble of magnitude val-ues. In spectral analysis, the result envelops thespectral values denoted by Ln, where L is thelevel exceeded n% of the duration of the mea-surement period.exponential averaging (also known asmoving averaging)The process of determining a power spectrumwh
20、ere the most recent record is most heavilyweighted to yield the arithmetic mean for thenumber of averages selected.filter averaging time (also known asfilter integration time)The length of time for averaging filtered data. Inmeasuring random data, the averaging time isinversely proportional to the e
21、rror in the power.For a given error or confidence limit, the re-quired averaging time is inversely proportionalto bandwidth; the governing relationship is com-monly referred to as the time bandwidth prod-uct.linear averaging (also known as stable,or power, averaging)The process of determining a powe
22、r spectrumwhere all records are weighted equally to yieldthe arithmetic mean for the number of averagesselected.maximum hold averaging (also knownas peak-hold or continuous peakaveraging)The process of determining a maximum signalvalue from an ensemble of magnitude values. Inspectral analysis, maxim
23、um hold averaging re-sults in the envelope of spectral amplitudes thatare not exceeded during the measurement ses-sion.bandwidth (also known as effectiveintegration bandwidth)The spacing between frequencies at which a band-pass filter attenuates the signal by 3 db.Note: In constant bandwidth applica
24、tions, band-width is often mistaken for the resolutionbandwidth. (The actual bandwidth is some-what larger than the resolution bandwidth, theratio depending upon the type of window weight-ing that is employed.)analysis bandwidthThe total relevant frequency range (for example,1 to 100 Hz), independen
25、t of type of analysis,resolution, or both.constant bandwidth (also known asnarrowband)A form of analysis or band-pass filter in whichthe bandwidth is constant and independent ofthe center frequency.constant percentage bandwidth (alsoknown as proportional bandwidth)A form of analysis or band-pass fil
26、ter in whichthe bandwidth is proportional to a center fre-quency. The most common forms of constantpercentage bandwidth are octave bands andone-third octave bands, defined in ANSI-S1.11(see section 2) as 69% and 23% of center frequen-cy bandwidth, respectively.integration bandwidthThe width of the f
27、requency band over which ameasurement is made or a mathematical oper-ation is performed. The amplitudes of vibrationare a direct function of the integration band-width used; i.e., the greater the bandwidth, thelarger the amplitudes.IEST-RP-CC024.1 INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY 7
28、Copyrightedmaterialresolution bandwidth (sometimescalled nominal bandwidth)The spacing between center frequencies.band-pass filterA frequency selective electronic circuit designed topass, that is either amplify or not attenuate, all datawithin a frequency range defined by a nonzerolower and finite u
29、pper limiting frequency.filtered frequency analysisThe continuous, simultaneous analysis of a signalin all relevant frequency bands. As a general rule,this process is performed with the use of parallelfilters. It is practical only for obtaining constant-percentage bandwidth spectra.Fourier transform
30、The process of transforming data from the timedomain to the frequency domain. Signal analyzersuse a fast Fourier transform (FFT) algorithm toaccomplish this transformation; hence, they areoften referred to as FFT analyzers.fast Fourier transform (FFT)A process of transforming data from the timedomai
31、n to the frequency domain. The algo-rithm computes discrete frequency componentsfrom discrete sampled time domain data.frequencyThe number of full cycle oscillations of vibrationper second, expressed Hertz (Hz).center frequencyThe center of the frequency span of a band-passfilter or frequency band.
32、ANSI-S1.11 (see section2) defines standard center frequencies for themore common forms of constant percentagebandwidth.frequency domainEquivalent representations of physical motion,wherein motion is quantified as a set of ampli-tudes as a function of frequency.frequency response functionThe ratio of
33、 the Fourier transform of the outputdivided by the Fourier transform of the input, asa function of frequency.natural frequencyThe frequency or frequencies at which a bodyvibrates as a result of its own physical character-istics (for example, mass and stiffness), whenthe body is excited.synthesized f
34、requency analysisAn analysis in which a constant percentagebandwidth spectrum is fabricated from two ormore constant bandwidth fast Fourier trans-form (FFT) spectra of frequency ranges thatdiffer by increasing orders of magnitude.leakageAn artifact of FFT analysis wherein energy appearsto spread thr
35、oughout the frequency domain, exhib-ited as the transfer of energy out of one resolutionline into other lines.linear spectrumSee rms spectrum.modal testingTesting that is performed to obtain the frequencyresponse functions of process tools or structures.These functions provide information about natu
36、ralfrequencies and characteristic shapes (modeshapes).noise floor (threshold)The voltage output of a transducer and associatedelectronics that is not related to motion of thetransducer. The noise floor (threshold) determinesthe minimum amplitude of motion that can bemeasured by a transducer-amplifie
37、r system.octaveThe interval between two frequencies having aratio of 2 to 1.one-third octaveThe interval between two frequencies having aratio of 1.26 to 1 (the cube root of 2).peak-to-peak valueThe absolute sum of positive and negative peakamplitudes.power spectrumThe mean squared amplitude (displa
38、cement, ve-locity, or acceleration) of a waveform as a functionof frequency. In denoting an acceleration powerspectrum, the units are expressed as gravitationalconstant squared, g2.power spectral density (PSD)The mean squared amplitude (displacement,velocity, or acceleration) per unit frequency of a
39、waveform. In denoting an acceleration PSD, theunits are expressed in g2/Hz.process toolVibration-sensitive equipment, including lithogra-IEST-RP-CC024.1 INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY8Copyrightedmaterialnarrowband vibrationVibration over a narrow frequency band; forexample, the a
40、mplitude of vibration between 1and 1.25 Hz (a 0.25 Hertz-wide band).periodic vibrationVibration characterized as a set of amplitudevalues that recur after a given interval known asthe period of motion.random vibrationVibration that is not periodic in nature. Becausethe term frequency cannot be rigor
41、ously appliedto random motion, the amplitude of randomvibration can only be specified in terms of aprobability function.simple harmonic vibrationVibration characterized by motion for whichthe amplitude is a sinusoidal function of time.transient vibrationVibration characterized by temporary sustained
42、motion with amplitudes that nominally startand end at zero.vibration thresholdThe point at which process tools can withstandvibration but still perform within acceptable limits.The vibration threshold may be expressed as anrms (linear) spectrum, or as single numerical val-ues of displacement, veloci
43、ty, or acceleration.window weightingApplying a profile or function on the time signalprior to doing the Fourier transform.zero-to-peak valueThe difference between zero and the maximumsignal level. In the case of simple harmonic motion,the zero-to-peak value is equal to the maximumsingle amplitude.4
44、BACKGROUND:OVERVIEW OF THE PROCESSOF ESTABLISHINGVIBRATION CRITERIAVibration criteria for sensitive process tools servetwo primary purposes:a) to provide quantitative bases for facility design;andb) to permit evaluation of the acceptability of thevibration environment in an existing facility. Aphy e
45、quipment, scanning electron microscopes,inspection equipment, and metrology equipment.resolution linesThe number of discrete linear data bands providedby a signal analyzer for a given measurement fre-quency band width. For example, for an 800-lineanalyzer measuring a frequency band width that is100
46、Hz wide, each line or data band has a resolutionequal to 100/800 or 1/8 Hz.root-mean-square value (rms)The amplitude of continuous motion or voltagewith equal energy, compared to the signal beinganalyzed. It is calculated by squaring all of thevalues, adding the squares together, dividing thetotal b
47、y the number of values to obtain the mean,and then taking the square root of the result. In thecase of simple harmonic motion, the rms value isequal to:22x Maximum single amplitudeNote: The rms value of random vibration, for ex-ample, is not equal to the rms value of simpleharmonic motion.rms (linea
48、r) spectrumThe rms (mean) amplitude (displacement, velocity,or acceleration) of a waveform as a function offrequency.root-sum-square value (rss)A value derived by squaring numerical quantities,adding them together, and then taking the squareroot of the sum.signal-to-noise ratioThe ratio between the
49、output signal of a transducerand its noise floor.time domainEquivalent representations of physical motion,wherein motion is quantified as a set of amplitudesas a function of time.vibrationThe alternating transfer of energy between its po-tential and kinetic forms, expressed as acceleration,velocity, or displacement of a physical object, andof which may be time-varying.broadband vibrationVibration over a broad frequency band; for ex-ample, the total amplitude of vibration between1 and 100 Hz.IEST-RP-CC024.1 INSTITUTE OF ENVIRONMENTAL SCIENCES AND TECHNOLOGY 9Copyrighted
