Proceedings of Insert Conference Abbreviation: Insert Conference Name Insert Conference Date and Location

Put Paper Number Here A TABLE TOP CNC/CMM TEACHING APPARATUS Allan D. Spence McMaster University Hamilton, Ontario, Canada

ABSTRACT Mechanical and Manufacturing Engineering Technology curriculums commonly include courses in Computer Aided Design and Manufacturing (CAD/CAM), and integration with Computer Numerical Control (CNC) machine tools and Coordinate Measuring Machines (CMMs). When a commercially sold apparatus is used for the laboratory experience, emphasis is too often on how to use the equipment, at the expense of understanding the underlying technology. This is a natural result arising from lack of student access to the internal software and electronics used. After graduation, however, students are expected to participate in design of such equipment, and therefore must be exposed to its operating principles. To address this need, a table top, open architecture CNC/CMM teaching apparatus was developed and integrated with an industry standard CAD/CAM software package. Accompanying curriculum was developed to serve undergraduate laboratories in Computer Numerical Control and Coordinate Metrology.

INTRODUCTION The need to expose Mechanical and Manufacturing Engineering and Engineering Technology students to CAD/CAM software and CNC/CMM manufacturing equipment operation is widely recognized. This is typically accomplished through some combination of composite laboratory and design project courses. Byrkett and Ettouney [1] discuss the evolutionary integration of a Computer Integrated Manufacturing (CIM) laboratory into an engineering education curriculum. The equipment was setup in separate Computer-aided Experimentation, Introductory Computer-aided and ComputerNumerical Control, and Flexible-Manufacturing System and CAD/CAM Laboratories. Students were given exercises on use

Harley L. Chan McMaster University Hamilton, Ontario, Canada

of AutoCAD for design, and SmartCAM for CNC machining. Senior courses, including a senior project, require more unstructured use of the facilities. Fang [2] presents a computer animation system for machining simulation. This system graphically displays a bar turning process on a computer monitor, including a simulated power meter. Students can “operate” the lathe by adjusting the speed and feed. The effects are observed in the animation, and the change in power. Fang promotes the system as an economical means of covering the subject material when faced with large class sizes and constrained laboratory facilities. Shiue et al [3] describe an Integrated Laboratory for Manufacturing Education (ILME) using modern Pro/Engineer CAD/CAM software and machine tools such as a Fadal VMC-15 CNC vertical machining center. The laboratory supports courses ranging from first year engineering graphics through senior projects. Proctor and Michaloski [4] describe an Enhanced Machine Controller (EMC) architecture developed at NIST. This architecture is now used on a few testbeds, and has been updated to include both DC motor and stepper motor support. The implementation interprets G-codes, and, using RealTime Linux [5], can operate a machine tool. Further details are available at the web site www.isd.cme.nist.gov/projects/emc. A CMM Dimensional Measuring Interface Standard (DMIS) interpretor is also available from NIST [6]. These implementations are open architecture, but are intended for larger equipment rather than teaching laboratory use. There is no integration with CAD/CAM systems, nor integration of CNC/CMM on a single apparatus. TORCOMP Systems Ltd. [7] commercially markets a teaching CNC machine. It is larger than a table top, does not support CMM education, and has a closed architecture. The above papers are representative of current curriculum content and available laboratory equipment. Commercial

1

Copyright © 2000 by ASME

equipment is purchased, or received as a donation. This minimizes the investment by faculty members and support staff, who do not have the time to design and develop custom, open architecture equipment. The corresponding curriculum exercises emphasize using the equipment. Although this is appropriate for courses on Product Design for Manufacturing, it does not adequately provide manufacturing students with the exposure to the applied science understanding of the equipment operating principles. For composite laboratories, there is a temptation to resort to “virtual laboratories”, or simulations, to reduce the cost of servicing large class sizes. These simulations, no matter how realistic they may be, cannot replace the tangible experience gained from experimenting with functioning equipment. At McMaster University, we use several simulation stations, combined with a small number of completely functioning equipment stations, for composite laboratory courses. Because the computer Graphical User Interface for all stations is identical, students can prepare their work on the simulation station, and then easily transfer to an equipment station. Custom designed, low power, low cost, open architecture, table top equipment is used. This provides a safe learning environment, with electronics and computer software that is openly accessible to the students for experimentation. Single instance, higher powered equipment is freed from composite laboratory use, and hence is available without interruption for project use. The custom equipment design will be made available, through the newly established Canadian-Design Engineering Network (C-DEN), to other educational institutions. The remainder of the paper describes a table top CNC/CMM apparatus that is used to teach solid modeler based CNC geometric tool path verification, Digital Differential Analyzer (DDA) based linear and circular interpolation, CMM touch trigger point data collection, and geometric feature data fitting. The following sections describe the system components, the CNC and CMM curriculum, and CAD/CAM/CNC/CMM integration. A discussion of ongoing improvements to this apparatus and other related projects is also included.

NOMENCLATURE a , b, c Linear interpolation X , Y , Z components (steps) Ap , Al Plane, line Singular Value Decomposition sample point matrices Linear table granularity (steps/mm) d Plane, circle sample i orthogonal error e ip , eci f Timer callback frequency (Hz) Ec Circle orthogonal sum of squares Feed rate (mm/min) F Circle data fitting convergence step index h i, j, k Circular interpolation X , Y , Z current position to center distances (steps) Linear interpolation path length (steps) L Number of sample points n p X , pY , pZ DDA addition registers

Fig. 1. CNC/CMM System Schematic

q X , qY , q Z rc rp

R s ip , sli , sci S touch , S pretouch S ip , Sli , S ci

S pm , Slm , S cm u, w

DDA quotient registers Circle best fit radius (mm) CMM touch probe radius (mm) Circular interpolation arc radius (steps) Plane, line, circle sample point i after translation by S pm , Slm , S cm CMM touch, pretouch point Plane, line, circle sample point i (after probe compensation) Plane, line, circle sample point mean

v p , vl T X ,Y, Z x c , yc

Linearized circle data fitting intermediate variables Plane, line unit normal vectors Timer callback period (s) Linear table axes Fitted circle center

δ x ,h , δ y ,h ε ic Εc ρ

Circle center correction vector components Linearized circle sample i error Linearized circle sum of squares Circle convergence criterion

SYSTEM COMPONENTS A schematic diagram of the complete CNC/CMM system is shown in Fig. 1. A 500 MHz Pentium III based Windows NT workstation, with 256 MB RAM and 20 GB of disk space, is at the heart. SDRC I-DEAS Master Series is used as the CAD/CAM and Post Processor software. A custom written Microsoft Visual C++ program is used to parse the CL-DATA and/or G-code files. A pipe based interface to the Spatial Technology Inc. ACIS solid modeler is used for CNC geometric verification. The stepper motor based machine tool is controlled in real time, in a separate thread, using the Windows NT multimedia timer. A National Instruments PCI-6503 digital I/O card provides the interface to the stepper motor amplifiers, limit switches, and CMM touch trigger probe. A variable speed Dremill tool is used as the CNC motor, spindle and chuck. Off the shelf carbide cutters are used for machining. The touch trigger probe system is a Renishaw PH1/TP2. An analog game joystick is used for manual positioning. The working volume of the machine is 100 mm ( X ) by 100 mm (Y ) by 50 mm ( Z ) . A photograph of the system is shown in Fig. 2. Full system specifications are shown in the Appendix. Additional details will be maintained at the web site www.mech.mcmaster.ca.

2

Copyright © 2000 by ASME

(a)

Fig. 2. CNC/CMM System Photograph

CNC CURRICULUM A milestone in automated manufacturing was the achievement of multiple axis coordinated motion using a stored NC program by the Parsons Corporation in 1951. Stepper motor driven implementations achieve coordinated motion using the Digital Differential Analyzer (DDA) or reference pulse method [8]. The linear interpolation DDA algorithm is schematically represented in Fig. 3. For our apparatus, the stepper motor resolution is 400 steps per revolution, and the lead screw advances 2 mm per revolution. The linear resolution is therefore d = 200 steps per millimeter, and hence each motor step advances the corresponding axis by 5 microns. For a user specified feed rate of F millimeters per minute, the timer callback frequency is f = dF / 60 Hz, and the callback period is T = 60 / dF seconds. Illustrating with the XY plane, a path of L steps is separated into components of a steps along the X axis, and b steps along the Y axis. Note that a and b are absolute values. The stepper motor direction is set by a separate electronic signal. The constant value a is loaded into the p X register, and the constant value b is loaded into the pY register. Both the q X and qY registers are initialized to zero. During each callback function execution, the value of the p X register is added to the q X register. The q X register

(b) Fig. 3. Linear DDA Algorithm: (a) Path Variables; (b) Algorithm Schematic

(a)

overflows when its value exceeds L = a 2 + b 2 . When this occurs, a pulse is sent to advance the X axis stepper motor by one step, and L is subtracted from q X . This architecture achieves an effective X axis step rate of fa / L Hz. The Y axis is handled similarly, achieving an effective Y axis step rate of fb / L . The total effective step rate is therefore the desired

f a 2 + b 2 / L = f Hz. Extension to add a Z axis involves

(b)

expressing the path length as L = a + b + c where c is the Z axis component, and including pZ and q Z registers.

Fig. 4. Circular DDA Algorithm: (a) Path Variables; (b) Algorithm Schematic

2

2

2

3

Copyright © 2000 by ASME

Fig. 6. Yoke Shaped Part Used for Inspection

Fig. 5. CMM Curriculum Apparatus Setup

S mp

A schematic representation of the circular DDA algorithm is shown in Fig. 4. Again the XY plane is used for illustration. The q X and qY registers are again initialized to zero. Overflow occurs when the register value exceeds the arc radius

R = i 2 + j 2 , where i is the X component from the start of the arc to the arc center, and j is the Y component from the start of the arc to the arc center. The difference from the linear interpolation case is that the p register values change. For example, consider a clockwise arc from the 9 o'clock to the 12 o'clock position. For this case, the initial values to load are p X = j and pY = i . Each time a step is issued in the X direction, the pY register is decremented by one. Each time a step is issued in the Y direction, the p X register is incremented by one. Similar situations hold for the other quadrants, and for counterclockwise motion. For arcs in the YZ or ZX plane, k is the Z component from the start of the arc to the arc center. Our curriculum requires fourth year undergraduate students to prepare for the laboratory by first studying a non real time MATLAB [9] graphical implementation of the linear and circular DDA algorithms. They are also provided with a Microsoft Visual C++ multimedia timer callback function that implements the linear DDA algorithm in real time. The laboratory assignment is to implement the circular DDA algorithm in real time. To verify correctness of the solution, a pencil is clamped in the Dremill chuck, and a pad of paper is clamped to the table. Because the equipment is low power, supervision during use is not required. Students attend the introductory laboratory demonstration at a scheduled time, but can complete the assignment on their own. They schedule a time with the Teaching Assistant to present their solution, and to carry out actual machining.

sip

vp

S

Fig. 7. Best Fit Plane Variable Definitions volume, is used for probe calibration. A small, yoke shaped part (Fig. 6), with plane, slot, and hole features, is secured to the table. Students learn Geometric Dimensioning & Tolerancing (GD&T) [10] in a prior course. Algorithms are based on those used in the NIST Algorithm Testing System (ATS) [11]. The laboratory assignments first require students to calibrate the probe using the datum sphere to obtain the effective probe radius rp . For all subsequent measurements, both the touch point S touch = ( X touch , Y touch , Z touch ) and the are pretouch point S pretouch = ( X pretouch , Y pretouch , Z pretouch ) recorded, and probe compensation is applied using the formula

S = S touch +

S touch − S pretouch S touch − S pretouch

rp

(1)

Next, students measure several points S ip = ( X ip , Ypi , Z ip ) (Fig. 7) on the top face of the yoke shaped part. The orthogonal least squares method is then used to find the best fit plane. First, the mean

S pm = CMM CURRICULUM For the CMM curriculum, the Dremill tool is replaced with a Renishaw PH1/TP2 touch probe system (Fig. 5). An analog game joystick provides easy manual operation. A 12 millimeter diameter datum sphere, mounted in the corner of the working

eip = (sip ⋅ vp ) vp

i p

1 n i ∑ Sp n i =1

(2)

is calculated. This defines a point on the best fit plane. The sample points are then translated to obtain their coordinates relative to the mean 4

Copyright © 2000 by ASME

s ip = S ip − S pm

(3)

Denoting the plane normal by the unit vector v p , the orthogonal distance from the sample point to the plane can be expressed using the vector dot product e ip = ( s ip ⋅ v p ) v p

(4)

The sum to be minimized is n

∑e i =1

i p

⋅ e ip

=

∑ 3s n

i =1

3 8 ∑ 3 s ⋅ v 8 3v ⋅ v 8 ∑ 3s ⋅ v 8 i p

8

⋅ v p v p ⋅ s ip ⋅ v p v p

n

=

i =1

2

i p

p

p

n

=

Fig. 8. Best Fit Circle Variable Definitions

2

i p

i =1

(5)

p

p

This can be written as a quadratic form v Tp A Tp A p v p , where

Ap = s1p

s 3p L s np

s 2p

T

and T is the transpose operator.

The extreme values of the quadratic form occur when v p is aligned with the eigenvectors of ApT Ap . Specifically, the best fit plane normal v p is the unit length eigenvector corresponding to the minimum eigenvalue of ApT Ap . In actual implementation, the equivalent problem of finding the eigenvalues and eigenvectors of Ap is solved, using the Singular Value Decomposition (SVD) [12]. The direction of v p is uniquely determined by again considering the touch and pretouch points. Together S pm and v p fully describe the orthogonal least squares best fit plane. Students next measure several sample points S li = ( X li , Yl i , Z li ) along the side face of the yoke shaped part. The orthogonal least squares best fit line is then determined. The mean 1 n

S lm =

n

∑

i =1

(6)

S li

Students finally measure a single point, and calculate the Homogeneous Transformation Matrix, based on the best fit plane, line, and point, to convert points coordinates between the Machine Coordinate Systems (MCS) and the Part Coordinate System (PCS). The vertical holes in yoke shaped part are then located using the orthogonal least squares circle algorithm. Denoting the sample points S ci = ( X ci , Yci ) , the mean

S cm =

s = S −S i l

( xci , yci ) = sci = S ci − S cm

(7)

The sum to be minimized in this case is n

∑s i =1

where

i l

eci = ( xci − xc ) 2 + ( yci − yc ) 2

1/ 2

− rc

(11)

where ( xc , yc ) is the estimated circle center location and rc is the estimated circle radius. The orthogonal least squares goal is to minimize the sum

E c = ∑ eci

2

(12)

i =1

Because of the square root term in Eq. (11), the problem is non-linear, and therefore must be solved iteratively from an initial approximation ( xc ,0 , yc , 0 ), rc , 0 . The method suggested by Anthony and Cox [13] is used. For nearly round circles, the error is defined as

n

− e ip ⋅ sli − e ip = ∑ sli ⋅ sli − vlT AlT Al vl

Al = sl1

(10)

The orthogonal error is defined as (Fig. 8)

n

m l

(9)

is calculated and the sample points are translated to obtain their coordinates relative to the mean

is a point on the line. The sample points are then translated to obtain their coordinates relative to the mean i l

1 n i ∑ Sc n i =1

(8)

ε ic = ( xci − xc ,0 ) 2 + ( yci − yc ,0 ) 2 − rc2,0

i =1

sl2

T

sl3 L sln .

The

eigenvector

(13)

and the sum of squares is written as

corresponding to the maximum eigenvalue defines the best fit line direction vector vl . Together S lm and vl fully describe the orthogonal least squares best fit line.

n

n

Ε c = ∑ ε ic = ∑ uci − w − 2 xci xc ,0 − 2 yci y c ,0 i =1

5

2

2

(14)

i =1

Copyright © 2000 by ASME

∂Ε c = 0; ∂w

∂Ε c = 0; ∂x c , 0

∂Ε c =0 ∂y c , 0

(15)

In matrix form



i =1 n

n

∑x i =1 n

!

"# ##  w ∑ x y # ⋅ 2x # !2 y ∑ y ## $

n

n

∑ xci

n

∑x

i c

i =1 n

∑ yci

i =1 n

i2 c

i =1 n

∑ yci ∑ yci xci i =1

i =1

i =1

i c

i c

c,0 c ,0

i2 c

 ∑ u "# "# ## = x u ## ∑ # $ y u ## !∑ #$ n

i c

i =1 n

i i c c

i =1 n

(16)

i i c c

i =1

After the value for w has been determined, the initial radius estimate is calculated from Fig. 9. I-DEAS Tool Path Graphical Display

rc ,0 = xc2,0 + yc2,0 + w

1/ 2

(17)

The minimum of Eq. (12) is now determined by iterating until the partial derivatives satisfy the non-linear conditions

∂E c = 0; ∂x c

∂E c =0 ∂yc

(18)

To improve the estimate, at step h , the second partial derivatives are also calculated, and the matrix equation

 ∂E "#  ∂ E ∂x # = ∂∂xE ∂E # ! ∂y #$ ! ∂y ∂x 2

c

c

c ,h

2 c ,h

2

c

c ,h

Fig. 10. Apparatus GUI / ACIS Verification

c

c,h

c ,h

"# ##  "# #$ ! $

∂ 2 Ec δxc ,h ∂xc ,h ∂yc ,h ⋅ 2 δyc ,h ∂ Ec 2 ∂yc ,h

is solved for the center correction vector δxc ,h

(19)

T

δyc ,h . The

improved center position estimate is

 x "# =  x "# − δx "# ! y $ ! y $ !δy $ c , h +1

c ,h

c ,h

c ,h +1

c ,h

c ,h

(20)

and the improved radius estimate is

rc ,h +1 =

1 n ∑ ( xci − xc,h +1 ) 2 + ( yci − yc,h +1 ) 2 n i =1

1/2

(21)

This iterative improvement procedure continues until the convergence condition

ρ > [ δ x c2,h + δ y c2, h ]1 / 2

Fig. 11. Machined and Assembled Clock 2

2

where uci = xci + y ci and w = rc2 − xc2, 0 − yc2,0 . estimate is found by simultaneously solving

The initial

(22)

is satisfied. The value for ρ is normally equal to the machine scale resolution. Finally, the circle center is translated back by Scm . Students implement this algorithm in MATLAB, and

6

Copyright © 2000 by ASME

compare to the answer calculated by the apparatus reference software, and to the yoke shaped part drawing.

CAD/CAM/CNC/CMM INTEGRATION For CAD/CAM/CNC integration, students first design a small clock face logo using I-DEAS. Tool paths are then planned, post processed, and graphically verified using the ACIS solid modeler based simulator. All of these steps can be completed offline, on any of 12 available workstations. When satisfied that the tool path is correct, the student moves to the workstation attached to the stepper motor apparatus. A Gravoply [14] multi-color laminated plastic blank is clamped onto the table, and actual machining is carried out. For typical designs, the machining operation requires 30-45 minutes. A small AA battery operated clock movement is secured to the machined face using double sided foam tape. Plastic tea cup hooks are attached to the rear lower corners of the face to complete a self standing desk clock. A graphical display of the I-DEAS tool path for the United Way logo is shown in Fig. 9. Figure 10 shows the Visual C++ based Graphical User Interface (GUI) to the ACIS simulator and stepper motors. A machined and assembled desk clock is shown in Fig. 11. A CAD/CAM/CMM integration assignment, highlighting B-spline surface measurement, is currently being developed.

DISCUSSION The DDA or reference pulse interpolation method was chosen because of its historical significance, ease of offline implementation in MATLAB, and because it would safely operate with Windows NT. An initial implementation in Windows 95 was abandoned because of irregular multimedia timer operation. The reference word interpolation method [15] is more commonly used with DC motor closed loop control systems. Use of this method on the teaching apparatus requires an unvarying loop closing period, and hence would have required the addition of hard real time NT kernel extensions, or use of RealTime Linux. Our curriculum includes a separate laboratory on analog and digital control systems in which the reference word interpolation method is studied. The CMM data fitting algorithms are also easy to implement offline in MATLAB. A laboratory on this subject matter is considered a valuable addition to the curriculum. Coordinate Measuring Machines are widely used, but poorly implemented data fitting mathematics has been reported as a source of concern [16]. A future extension would be to include the minimum circumscribed circle, and maximum inscribed circle, to support the Geometric Dimensioning & Tolerancing (GD&T) Maximum Material Condition (MMC) and Least Material Condition (LMC). Experience to date indicates that, so long as the subject material is clearly presented, students respond positively to learning how manufacturing automation equipment operates. A feeling of satisfaction accompanies mastery of this subject material that was previously lacking. When students now complete the CAD/CAM/CNC/CMM integration laboratory,

they do so with a confident understanding of the underlying operating principles. Development of the apparatus and accompanying curriculum has consumed significant resources in the form of final year design projects, technical staff assistance, graduate student effort, and faculty member supervision. A mandate of the Canadian-Design Engineering Network (C-DEN) is to share such efforts amongst participating Canadian educational institutions.

CONCLUSIONS This paper has presented a table top CNC/CMM apparatus that is used for several laboratories in manufacturing automation. The equipment was designed to be inexpensive, and open architecture. Unlike the purchase of commercially available equipment, the open architecture design allows students to more deeply investigate the equipment operating principles. To extend accessibility, duplicate offline stations are used for graphical CNC verification, and to implement the MATLAB portions of the assignments. It is intended that the design and curriculum be shared with other educational institutions through the recently established Canadian-Design Engineering Network.

ACKNOWLEDGMENTS The CNC/CMM teaching apparatus was developed as a part of several senior design projects. The participating students were Nicky Brunhuber, Yik Lung Chung, Marc DiCorrado, William DiDiodato, Michael Dunn, Robert Gawaziuk, Jason Lall, Joel Montgomery, Diane Newton, Leonard Oyama, Michael Potter, and Jeff Powell. Shad Valley summer students Joel Book and Matt Pazner provided additional assistance. The Faculty of Engineering and the Department of Mechanical Engineering provided funding to purchase the equipment. Robert Fleisig assisted with supervision of the senior design projects, and with programming the feature data fitting algorithms. Joe Verhaeghe and Ron Lodewyks assisted with electronics wiring and mechanical component machining. Omni-Tech CMM Services (Canada) Inc. donated the CMM probe head/touch probe.

REFERENCES 1. Byrkett, D. and Ettouney, O., “A Model to Develop and Incorporate a Computer Integrated Manufacturing Laboratory into an Engineering Curriculum”, Int. J. Engng Ed., Vol. 12, No. 4, pp. 272-281, 1996. 2. Fang, X.D., “Application of Computer Aided Animation of Machining Operations in Support of a Manufacturing Course”, Int. J. Engng Ed., Vol. 11, No. 6, pp. 435-440, 1995. 3. Shiue, Y.-S., Beard, B.B., Santi, M.L., and Beaini, J.E., “Integrated Laboratory for Manufacturing Education”, Int. J. Engng Ed., Vol. 15, No.1, pp. 51-57, 1999.

7

Copyright © 2000 by ASME

4.

5.

6.

7. 8.

9. 10. 11.

12.

13.

14. 15.

16.

Proctor, F. M., and Michaloski, J., "Enhanced Machine Controller Architecture Overview," NIST Internal Report 5331, National Institute of Standards and Technology, Gaithersburg, MA, 1993. Epplin, J., “Linux as an Embedded Operating System”, Embedded Systems Programming, Vol. 10, No. 10, Oct. 1997. available online at www.embedded.com. Kramer, T.R., Proctor, F.M., Rippey, W.G., Scott, H., “The NIST DMIS Interpreter Version 2”, NIST Internal Report 6252, National Institute of Standards and Technology, Gaithersburg, MA, 1998. TORCOMP Systems Ltd., www.torcomp.com. Koren, Y. and Masory, O., “Reference-Pulse Interpolators for CNC Systems”, Trans. ASME, J. Eng. Ind, Vol. 103, pp. 131-136, Feb. 1981. The MathWorks, Inc., MATLAB, version 5.3, Natick, MA. American Society of Mechanical Engineers, Dimensioning and Tolerancing, ASME Y14.5M-1994. C. Diaz and M.E.A. Algeo, “A Process for Selecting Standard Reference Algorithms for Evaluating Coordinate Measurement Software”, NIST Internal Report 5374, National Institute of Standards and Technology, Gaithersburg, MA, 1994. Press, W.H., Teukolsky, S.A., Vetterling, W.T., and Flannery, B.P., Numerical Recipes in C, 2nd Ed., Cambridge University Press, Cambridge, UK, 1992. Anthony, G.T. and Cox, M.G., “Reliable Algorithms for Roundness Assessment According to BS3730”, Proceedings of the NPL Meeting on Performance Verification of Coordinate Measuring Machines, National Physical Laboratory, Teddington, UK, 1988. Gravograph Ltd., London, UK, gravograph.co.uk. Masory, O. and Koren, Y., “Reference-Word Circular Interpolators for CNC Systems”, Trans. ASME, J. Eng. Ind., Vol. 104, pp. 400-405, Nov. 1982. Walker, R., GIDEP Alert No. X1-A-88-01, GovernmentIndustry Data Exchange Program, Washington, DC, 1988.

Spatial Technology Inc. ACIS V4.2 Machine Table: • For X and Y axis: • Stepper Motor: Superior Electric Model: M061-FD-301 Revolution: 400 steps/rev Max.Torque: 30 oz.in Voltage: DC 11.8 V Current: 0.44 A • Lead Screw: Parker Motion & Control Model: 102004S Travel: 100mm • For Z axis: • Stepper Motor: Superior Electric Model: MA61FS-62019 Revolution: 400 steps/rev Max. Torque: 60 oz.in Voltage: DC 10.1V Current: 0.5 A • Lead Screw: Parker Motion & Control Model: 102002S Travel: 50 mm Stepper Motor Amplifiers: • Current Amplification Integrated Circuit: Motorola Model: MC1314P • Stepper Drive: American Precision Industries Model: CMD-40 CNC Spindle: Dremel • Model: Moto-Tool 395 Type 5 Rotational Speed: 5,000-30,000 rpm Voltage: AC 117V Current: 1.15 A Touch Trigger Probe: Probe Head and Touch Probe: Renishaw PH1/ TP2

APPENDIX This appendix lists component details for the table top CNC/CMM apparatus. Specifications are subject to change, and will be maintained at the department web site www.mech.mcmaster.ca. Computer: • Hardware: CPU: Pentium III 500Hz RAM: 256 MB RAM Disk Space: 20 GB Analog Game Joystick: Gravis Digital I/O Card: National Instruments PCI-6503 • Software: Windows NT Workstation Version 4.00 SDRC I-DEAS Master Series 7.0 MATLAB Version 5.3 Microsoft Visual C++ 5.0

8

Copyright © 2000 by ASME