1、Errata to ASME B89 Technical Report I990 Parametric Calibration of Coordinate Measuring Machines On page 1, in third paragraph of para. 2.2, fourth line, Efforts revised to read Errors. On page 1, arrowheads added to Fig. 1. Revisions appear on the overleaf. On page 3, title of para. 5.0 revised to
2、read Squareness Measurement. On page 3, in last line of para. 5.1, diagonal square revised to read ball bar. On page 4, in Setup 1 of Fig. 3, left line representing straightedge revised to be continuous On page 11, in third paragraph of para. 8.3, third line, measurements and angle revised to read a
3、cross the probe. Revision appears on the overleaf. measurements, an angle. THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS Three Park Avenue, New York, NY 10016-5990 March 2002 L0070E FIG. 1 Probe FIG. 3 - - -7 /F-B87 TECH*REQT 70 I O757670 0081831 O E I I I ASME 889 TECHNICAL REPORT Parametric Cali br
4、ation of Coo rd i n ate Measuring Machines The American Society of Mechanical Engineers 345 East 47th Street, New York, N.Y. 10017 Date of Issuance: February 6, 1991 No part of this document may be reproduced in any form, in an electronic retrieval system or otherwise, without prior written permissi
5、on of the publisher. Copyright O 1991 by . THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS All Rights Reserved Printed in U.S.A. FOREWORD This Technical Report was written by Task Force L - Parametrics, ASME Working Group 889.1.12, Coordinate Measuring Machines. Methods for Performance Evaluation of Co
6、ordinate Measuring Machines, ANSI/ ASME B89.1.12M-1985, recognizes that niore-coiiiplex methods than those of tlie Stan- dard may be appropriate foi some applicatioiis. The intelit of this paper is to provide guidance to specification writers in those cases where tlie appropriatc iiiore-coinplcx iii
7、etliod is the method of parametric calibration. Since parametric calibration does hot assess the effects of inacliine dynaiiiics and probing, it should be considered a suppleiiient to B89.1.12 tests rather than a replacement. There are two types of appropriatc applications for parametric calibration
8、. The first includes cases where inethods of the B89.1.12 Standard do not give reasonable assurance machine geonietiy is correct. An example is the case where a short ball bar must be used to test a machine with one long axis. The second includes cases where there is a need to know details of machin
9、e geometry. An example is a system where the inacliine is used to position a photo array camera. Machines in a third category, those with parametric error correction, may generally be tested by methods of the Standard, aiid are outside the scope of this report. In many cases a partial parameric cali
10、bration may be appro- priate. Parametric calibration of a machine is measurement of its parametric errors. These are differences between actual and nominal positions of machine movable components. In tlie system of this paper, for each position of a three-axis machine there are twenty- one parametri
11、c errors. They may be mathematically coinbiiied to find the three axial componeiits of probe tip position error. Probe tip position error is one of several deter- minants of performance as measured, for example, in the B89.1.12 linear displacement accuracy aiid ball bar performance tests. The report
12、 is organized into several sections. The first is the scope. The second, dealing with machine geometiy, defines parametric errors in terms of an idealized example. The following five sections deal with measurements from which parametric errors ciln be calculated. The next section deals in a general
13、way with processing the measured data to determine the parainetric errors. The final two sections deal with terminology and references for further study. No attempt is made to be exhaustive. The intent, rather, is to pin down general principles, aiid to give the flavor of the work involved in a cali
14、bration. iii ASME 889 TECH*REPT 90 m 0759670 0083895 8 m ASME B89 COMMITTE Dimensional Metrolog iE IY (The following is the roster of the.Cornrnittee at the time of nnnrnvnl nf this TnPhni,-sl -r . -. OFFICERS 0. R. Taylor, Chairman F. G. Parsons, Vice Chairman P. D. Stumpf, Secretary COMMITTEE PERS
15、ONNEL A. M. Bratkovich J. B. Bryan A. K. Chitayat J. B. Dearn, Alternate W. E. Drews C. G. Erickson M. F. A. Fadl R. D. Fariey M. B. Grant R. B. Hook C. W. Jatho, Alternate F. L. Jones A. A. Lindberg E. G. Loewen T. Mukaihata D. Pieczulewski L. K. Reddig E. L. Watelet W. A. Watts J. H. Worthen R. B.
16、 Zipin PERSONNEL OF TASK FORCE L - PARAMETRICS W. L. Beckwith, Chairman J. B. Bryan R. P. Callaghan T. Charlton B. R. Taylor V PERSONNEL OF WORKING GROUP 1.12 - COORDINATE MEASUREMENT MACHINES (SCl/B89) R. B. Hook, Chairman W. P. Kiipatrick W. L. Beckwith, Jr., Secretary E. M. Kix, Jr. M. E. Algeo J
17、. S. Lifco:i M. Anderson E. G. Loeaen J. M. Baldwin L. Maggiano L. Best M. L. Majlak H. Bostrom W. McLendon J. B. Bryan D. R. McMurtry G. Burkett R. C. Meek R. P. Callaghan, Jr. F. M. Milis A. Campion K. E. Moser E. C. M. Caranfa D. M. Moyer T. Charlton, Jr. R. C. Newton R. T. Clark G. Nilsson A. C.
18、 Corbett J. T. Nilsson Y. Coue J. P. Olson D. J. Deer R. A. Olson J. Deis D. Ott R. Donaldson R. Parrillo W. E. Drews D. Fi. Porter D. T. Durboraw T. Posterick B. Edwards W. Randles J. Everhart W. H. Rasnick F. Farzan R. M. Fletcher S. Roufeh M. T. Gaie R. Shelton W. S. Gehner W. G. Clavik B. A. Gra
19、ham S. %fige D. M. Granrose W. J. Stegniuller M. B. Grant J. Stroili A. J. Griggs B. R. Taylor D. He W. B. Taylor G. P. Hegarty D. W. Thomas E. Helmel A. Traylor J. L. Henry K. Uibrich J. R. Hicks R. J. Hocken T. H. Hopp R. W. Horning J. Hurt R. M. Roterdam, Jr. R. C. Veale R. K. Walker R. W. Waller
20、 D. A. Wright CONTENTS Foreword . . iii Standards Committee Roster . v 1.0 Scope . 1 2.0 Machine Geometry 2.1 . General i . 2.2 Motion. Single Movable Coinponent . 2.3 Relationship Between Components 3.0 Positioning Error Measurements . 3.1 Generai 4.0 Straightness Error Measurement . 4.1 General .
21、4.2 Mechanical Straightedge 4.3 Taut Wire 4.4 Alignment Laser . 4.5 Laser Straightness Interferoineter 5.0 Squareness Measurement . 5.3 . Optical Square . 5.1 General . 5.2 Mechanical Square 5.4 Ball Bar 6.0 Yaw and Pitch Measurement . 6.1 General 6.2 Electrical Levels . 6.3 Autocolliintor . 6.4 . L
22、aser Angular Interferometer . 6.5 . Positioning Error Instrumentation 1 1 1. 2 2 2 2 2 2 3 3 3 3 3 3 7 7 7 7 7 7 7 7. 7.0 Roll Measurement . 9 7.1 General 9 7.2 . Electronic Levels . 9 7.3 . Straightness Instrumentation 9 9 8.1 Axes . 9 8.2 Measurement Lines . 11 8.3 Squareness . 11 8.4 Roll 11 9.0
23、Terminology 13 9.1 General . ; 13 9.2 Glossniy 13 14 8.0 Processing the Measured Data . 10.0 References for Further Study . vii ASME I389 TECH*REPT 90 0759670 0082898 3 = 889 REPORT PARAMETRIC CALIBRATION OF COORDINATE MEASURING MACHINES 1.0 SCOPE This report deals with coordinate measuring niit- ch
24、ines having three linear axes perpendicular to encli otlier. Such machines liave tliree iiiovable coinpo- nents. An example is the moving-bridge machine, Fig. 1. The movable coiiiponents are the bridge, car- riage, and rani. 2.0 MACHINE GEOMETRY 2.1 General Discussion is in two steps. The first cove
25、rs motion of a single movable component such as the carriage. The second covers relationship between motions of tlie three components. ipr=t Bridge Probe 2 2.2 Motion, Single Movable Component FIG. 1 The carriage of a moving-bridge inacliine is shown in Fig. 2. The carriage is designed to move in tl
26、ie X direction along a guideway on the bridge. It is as- sumed variations in forces acting on the carriage do not change its shape. The machine lias an X scale which measures mo- tion of tlie carriage in the X direction relative to the bridge. Vertical Let motion of the bridge and ram along their gu
27、ideways be prevented by means wliicli do not dis- tort the X guideway. Let the carriage be moved along tlie X guideway. Efforts in tlie actual motion are shown in Fig. 2. There are tliree rotational errors about the machine axes: yaw, pitch, and roll. Since the carriage does not change shape as it i
28、noves, these errors are tlie saine for all parts of tlie carriage. There are also three translational errors. Horizontal and vertical straightness are deviations from straight line motion. Positioning error is X scale reading mi- nus actual travel of tlie carriage froin tlie position of zero scale r
29、eading. Unless the three rotational errors are zero, the three translational errors depend on ehe straightness positioning error . FIG. 2 point on the carriage at which they are measured. The six X-dependant errors, three rotational and three translational, form a set of six X parametric errors. The
30、re are sets of Y and Z parametric errors determined in the saine manner and having the sanie characteristics. To cope with the dependance of translational er- rors on the point at which they are nieasurcd, tlie calibration system of this paper makes use of a probe reference point and a machine axis
31、system. The probe reference point is fixed to the ram in a defined, identifiable position. The origin of the machine axis system is fixed to the table in a defined, identifiable position. The machine axes are generally parallel to the machine guideways. It is usually assumed, as in this report, tlia
32、t para- metric errors for an axis depend only on position along that axis. This is only approximately truc. For example in the moving bridge machine of Fig. 1, loads on the left and right bridge legs, and on the Y guideways, depend on X location of the carriage. De- flections of the guideways depend
33、 both on load and Y position. Thus Y parametric errors are functions of both X and Y. Position of the Y axis should be chosen so that Y parametric errors, determined as functions at Y only, are the most meaningful. Exact definitions of parametric errors according to the system of this report are bes
34、t illustrated by means of an idealized example. Steps in measuring yaw, pitch, and roll are the same. The probe reference point is placed at the nia- chine origin. Motions of the Y and 2 movable com- ponents are prevented by means which do not distort the X guideway. Gaging is set up to measure angu
35、lar motion of the ram relative to the table. Gage readout is zeroed. The iiiacliine is moved to predetermined X positions, and X scale gage readings are recorded. The recorded gage reading for a position is tlie par- ametric error for that position. Y and Z yaw, pitch, and roll are measured in a sim
36、ilar manner. For determination of X horizontal or vertical Straightness, tlie probe reference point is placed at the origin, and Y and 2 motions are prevented, as for X angular errors. A reference straightedge (nie- chanical, optical, etc.) is mounted to the table with its reference line approximate
37、ly along the X axis. A gaging device is mounted o11 the ram at the probe reference point to read against the straightedge. The machine is moved to predetermined X positions, and X scale and gage readings are recorded. A root- mean-square best fit line through the measured data is calculated. For eac
38、h X position, straightness is re- corded gage reading ininus corresponding coordinate 2 of the best-fit liiic. Y and Z straiclitness arc deter- mined in a similar manncr. For dctermiiintiun of X positioniiig ci-ror, the probe i-eferciicc point is placcd at tlic oi.igin, and Y and Z motions are prcvc
39、iitcd, ;is foi- X angular errors. Gaging is set up to measure X travcl of tlic ram rel- ative to tlie tiilc. Tlic inricliine is moved to prcdc- tcrminctl X positions. X scale ancl gagc readings arc recorded. For any X position! X travel error is scale reading ininus gagc rcacliiig. Y aiid Z position
40、ing er- ror arc dcterniinctl in the saiiie niaiinci-. 2.3 Relationship Between Components Relationships between coiiipoiicn t motions are de- scribed by three parametric errors known ;is square- nesses. XY squarcness is angle bctwccn X and Y best-fit lines, as vicwcd in the Z direction, minus a betw
41、een X and Z best-fit lines. YZ squareiiess is a sitnilnr relationship bctLvccii Y aiid Z bcst-fit lines. right anglc. xz s(u1rcIlcss is ;1 siiiiilar rclationsliip 3.0 POSITIONING ERROR MEASUREMENT 3.1 General Common gages for incasuring positioning errors are the step gage, lasci- intcrferoiiicter.
42、and line scale. Proper USC of the step gage and lasci. interferometer are discussed in the BS9.1.12 Standard, Section 5.1. Use of ;I prccisioii line scalc aiid inicroscopc is liiii- ited to low-accuracy iiiacliiiics. 4.0 STRAIGHTNESS MEASUREMENTS 4.1 General Com mon ni et li o cl s of s t rii i g li
43、 t n ess in e asu re ni en t s make use of the niechanical Straightedge, taut wire, aiignnient lasci-, and laser interferometer. 4.2 Mechanical Straightedge Meclianical straiglitcclgcs arc usually uscd only for sin ;i I I in acli i n es. TIi e s t ra i g l i t c cl sc is a p 11 ros i ni at c I y ali
44、gned parallel to ai: axis guideway atid measured with a ineclianical or clectronic indicator. Measure- ments may be corrected by means of straightedge calibration, but since calibrations are always suspect, a better inctliod is to rcvcrse tlic straiglitcdge. This is done by rotating the Straightedge
45、 1SO deg. about ASME BA9 TECH*REPT 90 O 0759b70 O081900 8 O its long axis and re-measuring. If corresponding val- ues of the two sets of readings are suitably averaged, straightedge errors cancel out. Typical setps are shown in Fig. 3. Whatever setups are used should be carefully analyzed to determi
46、ne the correct method of averaging. For example in Fig. 3, a positive gage reading in Setup 1 corresponds to a positive straight- ness error. In Setup 2 where the straightedge and gage head are reversed, a negative gage reading cor- responds with a positive error. Therefore the method of averaging i
47、s to subtract measurement of Setup 2 from corresponding measurement of Setup 1, .and divide the results by 2. For larger machines the straightedge must be staged. The dilemma with staging is that a large over- lap of gage positions tends to improve relative gage alignments, but more gage positions a
48、re required. It is veiy difficult to make valid measurements on a large machine. 4.3 Taut Wire Taut wire measurements, shown in Fig. 4, are often used for large machines. The wire is stretched along an axis-direction line, and measurements are made with a microscope mounted to the ram. Vertical stra
49、ightness measurements for a horizontal axis are usually impractical because of the difficulty of mak- ing an accurate catenary correction. 4.4 Alignment Laser The alignment laser, shown in Fig. 5, uses the cen- ter of a laser beam as the straightness reference line. The laser is mounted on the machine table. Lateral deviation of the ram from the laser beam center is sensed by a quadrant or lateral effect sensor on the ram. A light filter at the sensor may be needed to reduce the effect of ambient light. Since the laser beam tends to wander somewhat due to mechanical distortions,
