Develop procedures to determine geometric measurement errors after work-piece machining Dirk Beger,...
-
Upload
dina-mathews -
Category
Documents
-
view
216 -
download
0
Transcript of Develop procedures to determine geometric measurement errors after work-piece machining Dirk Beger,...
Develop procedures to determine geometric measurement errors after work-piece machiningDirk Beger, Lisa Groos, Klaus Wendt
TIM Workshop, London
5th November 2014
Page 2 of 27
Traceable in-process dimensional measurement
Objective: traceability of dimensional 3D
measurements on machine
tools
Requirement: reliable and traceable
measurement uncertainties;target accuracy
5 µm/m
Challenge: harsh environmental conditions e.g. no air-conditioning
Methods: use of • different material standards• different procedures• climate simulation chamber
Source: renishaw.com
Figure 3 Prismatic shapes
Page 3 of 27
Benefits of on-machine measurement
finishmachining
on-machinemeasurement re-machining ?
dismountworkpiece
remountworkpiece
measuringroom
workpiececoordinate setting
workpiececoordinate setting
Challenge: measurement under shop floor conditions complex determination of measurement
uncertainty
Benefit: shortened measurement duration reduction of energy costs (no expensive air-conditioning)
Target accuracy: 5 µm/m
Page 4 of 27
Procedures
• volumetric error mappingmapping of geometric errors across the entire working volume (generally requires a mathematical error model)
• verification of MPEvalidation of error compensation and MPE compliance test (ISO 230-6, ISO 230-10)
• task-specific error mapping and correctiondetermination of size, form and position errors of shape measurements / typical geometric features (generally does not require a mathematical error model)
Source: Hüser et al., 1996
x
y
zD3 D4
D1 D2Source: Modified from ISO 230-6
Page 5 of 27
Procedures
• task-specific measurement uncertaintydetermination of measurement uncertainty for shape and form measurements / typical geometric features
• fitness-of-purposeassessment of the capability of a machine tool to measure geometric features with sufficient small accuracy compared to the required manufacturing tolerance
lower specification limit
upper specification limit
work piece tolerance
non-conformity
uncertaintyregion
uncertaintyregion
non-conformityconformity
measurementuncertainty
measurementuncertainty
measurement uncertainty
operator
machine tool
probing system
measuring method
environment workpiece
Page 6 of 27
Correlation: procedure, standard and error
procedure materialstandards
error sources
volumetric error mappingverification of MPE
geometric, thermo-mechanical
task-specific error mappingtask-specific measurement uncertainty
+ dynamic forces+ motion control+ control software
fitness-of-purpose++ loads++ heat
Source: eµmetron
Figure 3 Prismatic shapes
Page 7 of 27
Geometric errors of linear and rotary axes – ISO 230-1
6 error motions
EAX roll
EBX yaw
ECX pitch
EXX positioning
EYX vertical straightness
EZX horizontal straightness
+ 3 location / orientation errors
6 error motions
EAC tilt
EBC tilt
ECC angular positioning
EXC radial
EYC radial
EZC axial
+ 5 location / orientation errors
Page 8 of 27
Determination of machine tool errors using a hole plate
thermo-invariant Zerodur® hole plate
• hole plate made of Zerodur®
• mapping of error motions and location/orientation errors (ISO 230-1)
• volumetric errors are determined for
temperatures between 15° and 45° (use of mobile climate simulation chamber)
Page 9 of 27
Determination of machine tool errors using a hole plate
• systematic evaluation of the geometric errors based on measurements of a hole plate in four positions
• developed for coordinate measurement machines at the PTB
• more information (e.g.):
A reference object based method to determine the parametric error components of coordinate measuring machines and machine tools
E. Trapet and F. WäldeleMeasurement Vol. 9 No. 1, pp. 17-22 (1989)
P osition 1
P osition 2
P osition 3
P osition 4
Linear axes subsystem:
Page 10 of 27
Determination of machine tool errors using a hole plate
Rotary axis subsystem:
x
y
Approach:• measure coordinates of at least 3 points
Assumption: • errors of linear axes already corrected
Page 11 of 27
Determination of machine tool errors using a hole plate
Rotary axis subsystem:
x
y
Approach:• measure coordinates of at least 3 points
• rotation around C-axis and measurement of the chosen points
Assumption: • errors of linear axes already corrected
Page 12 of 27
Approach:• measure coordinates of at least 3 points
• rotation around C-axis and measurement of the chosen points
• description of the rotation with homogeneous transformation matrices yields:
Determination of machine tool errors using a hole plate
Rotary axis subsystem:
ZCACBC
YCACCC
XCBCCC
EyExE
EzExE
EzEyE
z
y
x
• calculation of errors by a Gaussian fit (method of least-squares)
x
y
Assumption: • errors of linear axes already corrected
Page 13 of 27
Determination of measurement uncertainty
Approach:
• use of Monte-Carlo-Simulation within a virtual machine tool → expanded measurement uncertainty U
(of type B according to GUM)
Advantage:
• determine smallest measurable tolerance Tmin in advance via
G
UT min
G: limiting value specified by user
Page 14 of 27
Correlation: procedure, standard and error
procedure materialstandards
error sources
volumetric error mappingverification of MPE
geometric, thermo-mechanical
task-specific error mappingtask-specific measurement uncertainty
+ dynamic forces+ motion control+ control software
fitness-of-purpose++ loads++ heat
Source: eµmetron
Figure 3 Prismatic shapes
Page 15 of 27
Task-specific error compensation – material standard
• close to zero thermal expansion (made from Zerodur®)
• specifications:• form errors< 1 µm• roughness < 0.2 µm• position and size of features will be
calibrated on CMM
• measurement tasks:• hole• flat surface• coaxiality
Traceability for tactile measurements
Traceability for optical measurements
• extremely low coefficient of thermal expansion (made from Invar)
• specifications:• plated surface• form errors < 5 µm• position and size of features will be
calibrated on CMM, accuracy M 2 µm
• measurement tasks:• hemisphere, cylinder, cone• flat and perturbed surfaces
Figure 3 Prismatic shapes
Page 16 of 27
Task-specific error compensation – concept
Approach:
Calibrated artefact
• acquire local error vectors by measuring a calibrated artefact (e.g. Zerodur® Multi-Feature-Check or a workpiece replica)
Page 17 of 27
Task-specific error compensation – concept
Real workpiece
• measure real workpiece using tactile or optical probe
• use the local error vector to correct the points measured point by point
• evaluation of the corrected points
Approach:
Calibrated artefact
• acquire local error vectors by measuring a calibrated artefact (e.g. Zerodur® Multi-Feature-Check or a workpiece replica)
Page 18 of 27
Task-specific error compensation – concept
Real workpiece
• measure real workpiece using tactile or optical probe
• use the local error vector to correct the points measured point by point
• evaluation of the corrected points
Approach:
Calibrated artefact
• acquire local error vectors by measuring a calibrated artefact (e.g. Zerodur® Multi-Feature-Check or a workpiece replica)
Advantageno specific geometric error model of
the machine tool is required
Page 19 of 27
Task-specific error compensation
DesignCAD model
Manufacture5-axis machine tool
Measuretask-specific errors
Software interfaceAdjustment of manufacturing
parameters in numerical control for on-line error correction
Page 20 of 27
Task-specific error compensation
DesignCAD model
Manufacture5-axis machine tool
Measuretask-specific errors
Software interfaceAdjustment of manufacturing
parameters in numerical control for on-line error correction
Adjust CAD model
Page 21 of 27
Correlation: procedure, standard and error
procedure materialstandards
error sources
volumetric error mappingverification of MPE
geometric, thermo-mechanical
task-specific error mappingtask-specific measurement uncertainty
+ dynamic forces+ motion control+ control software
fitness-of-purpose++ loads++ heat
Source: eµmetron
Figure 3 Prismatic shapes
Page 22 of 27
Fitness-of-purpose
within tolerance
uncertaintyregion
region of conformity
lower specification limitspecification phase
verification phase
region of non-conformity
region of non-conformity
work piece toleranceout of tolerance out of tolerance
incr
easi
ng m
easu
rem
ent
unce
rtai
nty U
uncertaintyregion
upper specification limit
1.03.0 GT
Ug
Illustration adapted from ISO 14253-1
Page 23 of 27
Assessing fitness-of-purpose – concept
Experimental determination of measurement uncertainty by
workpiece replica material standard (WR-MS) according to ISO 15530-3:
Manufacturing of WR-MS on machine tool
Repeated measuring of WR-MS on machine tool
Calibration of WR-MS on a CMM
Additional error sources:Loads and internal
heat sources
Page 24 of 27
Assessing fitness-of-purpose – concept
Experimental determination of measurement uncertainty by
workpiece replica material standard (WR-MS) according to ISO 15530-3:
Manufacturing of WR-MS on machine tool
Repeated measuring of WR-MS on machine tool
Calibration of WR-MS on a CMM
Additional error sources:Loads and internal
heat sourcesMean value + standard
deviation of size, form or position measurements
ypu
Reference result for size, form or position +
calibration uncertainty ucalu
calx
Page 25 of 27
Assessing fitness-of-purpose – concept
Experimental determination of measurement uncertainty by
workpiece replica material standard (WR-MS) according to ISO 15530-3:
Manufacturing of WR-MS on machine tool
Repeated measuring of WR-MS on machine tool
Calibration of WR-MS on a CMM
ISO 15530-3
Additional error sources:Loads and internal
heat sources
Fitness-of-purpose: U/T ≤ G
Mean value + standard deviation of size, form or
position measurements
ypu
Reference result for size, form or position +
calibration uncertainty ucalu
calx
Additional uncertainties uw
from manufacturing process
Systematic error: calxyb
2222 buuukU wpcal Expanded measurement uncertainty:
Page 26 of 27
Conclusions
• Geometric errors of machine tool axes can be systematically mapped using a hole plate for instance, which allows to assess the task-specific measurement uncertainty by a virtual CMM / MT for instance
→ determination of smallest measurable tolerance
• Task-specific error mapping allows improving manufacturing accuracy
→ by adjustment of manufacturing parameters
• Measurement uncertainty can be experimentally determined by repeated measurements of calibrated artefacts
→ fitness-of-purpose
Ensure traceability of dimensional 3D measurements on machine tools
for quality assurance.
Physikalisch-Technische BundesanstaltBundesallee 10038116 Braunschweig, Germany
Dirk Beger, Lisa Groos, Klaus WendtWorking Group 5.32Coordinate Measuring Machines
Phone: +49 531 592-5349E-mail: [email protected]
As of: 10/2014