US 4,152,765 AGrant
Programmer Unit for N/C Systems
Issue Date:1979-05-01
•13 Claims
•20 Drawing Sheets
Abstract
A programmer unit includes a keyboard through which an operator may enter data which is responsive to questions displayed on an alpha-numeric display. The programmer automatically leads the operator through a series of steps in which a part is defined as a series of composites and stored in an (L) array. This definition of the part is then converted to a series of move commands which are stored sequentially in an (O) array. Cutter offset and machine codes are added to the data stored in the (O) array to form a part program which may be read out and stored on punched tape or magnetic tape.
Metadata
Inventor
- John M. Weber
Application Information
Application Number:US 8335095
Filing Date:1977-09-15
Priority Date:1977-09-15
Art Unit:236
Classifications
IPC:
G05B 1942
Field of Search:
364318107;118;120;474567;568;569;570
Patent Drawings (20 sheets)
Description
Background of the Invention
The field of the invention is numerically controlled machine tools, and more particularly, the development of part programs for numerical control systems on such machine tools.
Numerical control systems direct machine tools such as milling and drilling machines through a series of moves to machine a part. The desired series of moves form a part program which is input to the numerical control system. Such part programs may be stored on a magnetic or punched tape or they may be coupled through a communication link from a host computer in a DNC system. The format and coding of part programs for use on commercially available numerical control systems is standardized and thus as a practical matter a machine for developing such a part program must generate its work product in one of the standard forms.
Prior systems for generating part programs are expensive, time consuming and require the skills of a trained computer programmer. Such systems employ high level languages such as APT or UNIAPT to define the part which is to be machined. The operator must be skilled in the use of the language in order to formulate the statements which properly define the geometry. Following this, the developed program must be pre-processed and then post processed to convert the statements into a standard part program format and code. Such processing usually requires the use of a large general purpose digital computer which is expensive and which may involve considerable time. If an error is made, or if changes in the program are desired, the entire process usually must be repeated.
Prior systems for generating part programs are particularly impractical for small machine shops which cannot justify the cost of a general purpose digital computer or the hiring of a skilled programmer. As a result, the use of numerically controlled machine tools at such establishments has been limited. Also, even when the equipment and skilled programmer are present, the cost and time required to generate a part program may more than offset the advantages of making the part on a numerically controlled machine. This is often the case when only a few of the same part are to be made, in which case it is more practical to employ a manually operated machine.
Summary of the Invention
The present invention relates to a programmer unit for interactively developing a part program for a numerical control system. More specifically, the programmer unit includes a keyboard and display for interfacing with an operator; means for storing an (L) array and an (O) array; a plurality of selectable composite definition means, each associated with a unique geometric shape and including means for outputting data to the display and inputting responsive data from the keyboard, which inputted data includes a set of dimensions that define the associated unique geometric shape; and means for storing the set of dimensions in the (L) array along with its associated composite number.
A definition of the geometric shape of the part is formed by a series of composite numbers and their associated dimensions which are stored in the (L) array. This geometric definition is converted to a part program by selectable composite processor means which is responsive to the composite numbers stored in the (L) array. The composite processor means converts a set of dimensions stored in the (L) array to sets of elements stored in the (O) array which define one or more moves to be made by a machine tool to form the part. The contents of the (O) array can be outputted directly to the numerical control system on a machine tool or it can be stored on punched or magnetic tape for later use.
An underlying concept of the present invention is that most parts which are machined with numerically controlled machine tools can be viewed as a series of relatively simple concatenated geometric shapes. It is another aspect of the invention, however, that certain of such geometric shapes can be readily defined by dimensions which are commonly available on drawings of parts to be machined. Each of the composite definition means in the invented programmer unit is associated with one of such geometric shapes and when one is selected by the operator, the programmer unit requests the operator to input the set of dimensions, or parameters, necessary to specifically define the selected geometric shape. Such dimensions may be, for example, the end point of a straight line segment, the center and radii of circular arcs, or merely the radii of a circular arc. The part can thus be constructed of a series of selected geometric shapes and each of these geometric shapes may be precisely defined by readily available information.
The resulting definition of the part is stored in the (L) array and the programmer unit then automatically selects and employs the appropriate composite processor means to calculate the elements of a series of moves which will machine the part. Such moves may include a straight line from the present position of the machine tool to an end point defined by X and Y coordinate values, or it may include a circular move about a center defined by X and Y coordinates to an end point defined by X and Y coordinates. The elements of the series of moves which define the entire part to be machined are stored in the (O) array and in some cases this data may be employed directly to control the machine. More typically, however, the move elements are recalculated to include cutter offset.
A general object of the invention is to facilitate the development of part programs for numerically controlled machines. Because the programmer unit interacts with the operator and guides him through the steps of developing the part program, highly trained operators are not necessary. The programmer unit interrogates the operator as to which composite number is to be selected next, and when selected, the programmer sequentially indicates to the operator which dimensions must be entered to define the selected geometric shape.
Another general object of the invention is to provide a means for rapidly developing a part program. Most parts are easily broken down into a series of geometric shapes which are recognized by the programmer unit. The operator selects the composite number associated with each of the geometric shapes and enters the requested dimensions. In a matter of seconds the programmer unit automatically calculates the elements of the (O) array from the elements of the (L) array and then, if desired, the part program can be tested on an X-Y plotter. If needed, cutter offset can be entered and the programmer unit automatically recalculates the elements of the (O) array to take this into consideration.
The foregoing and other objects and advantages of the invention will appear from the foregoing description. In the description reference is made to the accompanying drawings which form a part hereof, and in which there is shown by way of illustration a preferred embodiment of the invention. Such embodiment does not necessarily represent the full scope of the invention, however, and reference is made to the claims herein for interpreting the breadth of the invention.
Brief Description of the Drawings
FIG. 1 is a perspective view of the invented programmer connected to a numerical control system,
FIG. 2 is an electrical block diagram of the programmer,
FIG. 3 is an electrical block diagram of the I/O control unit which forms part of the programmer of FIG. 2,
FIG. 4 is schematic diagram of the system software of the programmer of FIG. 1,
FIGS. 5-13 are flow charts of the programmer main routine which is stored in a memory which forms part of the programmer of FIG. 2,
FIG. 14 is a drawing of an exemplary part for which a part program may be developed using the present invention,
FIG. 15 is an illustration of the five composites which may be selected to define a part,
FIG. 16 is a schematic illustration of the example part of FIG. 14 broken into a series of the composites of FIG. 15,
FIG. 17 is a schematic representation of the (L) array which is stored in the programmer of FIG. 2,
FIG. 18 is a schematic representation of the (O) array which is stored in the programmer of FIG. 2.
FIG. 19 is an illustration of the eleven subtypes into which the five composites of FIG. 15 are broken in order to facilitate calculations, and
FIG. 20a and 20b are illustrations of the variables involved in calculating elements of the (O) array.
Description of the Preferred Embodiment
Referring to FIG. 1, the numerical control programmer unit includes a keyboard unit 1 for entering information into and controlling the operation of the programmer. It also includes a magnetic tape cassette unit 2 for recording information stored in the programmer and for loading recorder information into the programmer. The programmer unit also includes a solid state output display unit 3 which provides visual indication of alpha-numeric data stored within the programmer. All of these elements are contained within a single housing 4 which may be easily transported to a selected site.
The numerical control programmer unit may, for example, connect to the DNC interface terminal on a numerical control system 5 through a cable 8. The program developed with the use of the programmer unit may thus be directly transferred to the numerical control system 5 which uses it to control a multi-axis machine tool 7 through a cable 6. As will be described in more detail hereinafter, other input and output devices may be connected to the programmer unit to input data thereto and to store the program which is developed. The developed program may be stored on punched tape, for example, which may be used on the tape reader of most commercially available numerical control systems.
Referring to FIG. 2, the programmer unit operates in response to instructions stored in a memory 10. The memory 10 is an MOS-type semiconductor memory which stores 16-bit words that are separately addressable through a memory address register 11. The addressed word in the memory 10 may be read out and stored in a 16-bit memory data register 12, or conversely, a 16-bit word in the memory data register 12 may be written into an addressed line of the memory 10. The registers 11 and 12 are recirculating 16-bit serial shift registers each having a serial input terminal connected to a T-bus 13 and each having a serial output terminal connected to an S-bus 14. Data is coupled between the memory 10 and the other elements of the programmer through the T-bus 13 and the S-bus 14.
The programmer also includes four registers which connect to the T-bus 13 and an R-bus 19. These are identified as the P register 15, the E register 16, the A register 17 and the B register 18. The P, A and B registers are recirculating 16-bit serial shift registers having their serial inputs connected to the T-bus 13 and their serial outputs connected to the R-bus 19. The E register 16 is a four-bit shift register having its input connected to the T-bus 13 and its output connected to the R-bus 19. The P register 15 serves as the program counter and the A and B registers 17 and 18 serve as the accummulator registers. The E registers serves as a serial extend register.
The T-bus 13, the S-bus 14 and the R-bus 19 connect to an arithmetic logic unit 20. The arithmetic logic unit 20 performs one-bit serial binary arithmetic on data received through the S-bus 14 or data received through the R-bus 19 and it may also perform logic operations on data received through the buses 14 and 19. The arithmetic and logic operations performed by the arithmetic logic unit 20 are determined by microinstructions which are received through a control bus 21 from a microprocessor 22. Results of arithmetic or logical operations performed by the arithmetic and logic unit 20 are transferred serially through the T-bus 13 to any one of the above described registers.
The operation of all elements in the programmer is determined by microinstructions generated on the control bus 21 by the microprocessor 22. The microprocessor 22 is in turn responsive to 16-bit macroinstructions which are stored in an instruction register 23. The instruction register 23 is a recirculating 16-bit shift register having its serial input connected to the T-bus 13 and its serial output connected to the R-bus 19. Its sixteen parallel output terminals connect to the microprocessor 22.
A basic machine operation consists of loading the contents of the P register 15 into the memory address register 11 and reading the macroinstruction on the addressed memory line out of the memory 10 and into the memory data register 12. The macroinstruction is then shifted to the instruction register 23 where it is applied to the microprocessor 22. The microprocessor 22 responds by issuing a series of microinstructions on the control bus 21 which operate the system elements to carry out the functions indicated by the macroinstruction. After the macroinstruction is executed, the program counter (P register 15) is incremented and the cycle repeats to execute the next macroinstruction. For a more detailed description of the structure of the programmer unit and the manner in which it operates to execute instructions, reference is made to U.S. Pat. No. 4,012,725 which issued on Mar. 15, 1977 and is entitled "Programmable Calculator."
Data is coupled to and from the various I/O devices associated with the programmer through an I/O control unit 25. Referring particularly to FIGS. 2 and 3, the I/O control unit 25 has a serial input 26 which receives output data from the T-bus 13 and a serial output terminal 27 which connects to the S-bus 14. The input terminal 26 and the output terminal 27 are associated with a universal 16-bit I/O shift register 28 into which data may be transferred either bit-serially from the T-bus 13 or in parallel from an input data bus 29. Data may be transferred from the I/O register 28 either bit-serially to the S-bus 14 or in parallel to an output data bus 30.
The input data bus 29 and the output data bus 30 couple to a variety of I/O devices. Both buses 29 and 30 connect to the keyboard 1 and the tape cassette 2 and the output data bus 30 couples to the alpha-numeric display 3. Other peripheral devices including an X-Y plotter 40, a tape punch 41 and a digital interface circuit 42 connect to the input data bus 29 and the output data bus 30.
An interrupt control circuit 31 is connected to the keyboard 1 through a cable 32 and when a key is depressed, an interrupt request signal is generated through a lead 33 to the microprocessor 22. At the completion of the macroinstruction then being executed, the microprocessor 22 calls up an interrupt service routine stored in the memory 10. The interrupt service routine inputs the character from the keyboard 1 and then returns to the interrupted routine.
Associated with the alpha-numeric display 3 is a display control circuit 34 which has an input connected through a lead 35 to the fourth-bit output terminal on the instruction register 23. The display control circuit 34 connects to the alpha-numeric display 3 through a lead 36 and when data is to be outputted to the alpha-numeric display 3, it is enabled by the display control circuit 34. Such an operation is performed in response to an I/O macroinstruction stored in the instruction register 23. The I/O macroinstruction includes an operation code which indicates that data is to be outputted to the alpha-numeric display 3 and an operand which identifies the source of data to be outputted. The microprocessor 22 responds by generating microinstructions through the control bus 21 which couple the identified data through the T-bus 13 into the I/O register 28 and then on to the output data bus 30.
The digital interface circuit connects to the cable 8 which couples to the numerical control system 5 and it operates to provide an EIA-RS-232C compatible interface with the cable 8. Part program data developed with the use of the programmer unit may thus be coupled directly to the memory of the numerical control system 5 through the cable 8. This capability is particularly useful when the programmer is used in conjunction with numerical control systems which provide for the storage in their memory of entire part programs. The cable 8 may also connect to a digital printer or a host computer in a DNC system.
If direct coupling between the programmer unit and the numerical control system 5 is not desired, or not possible, the part program developed with the use of the programmer unit may be tested on the X-Y plotter 40 and then outputted to the tape punch 41. Operation of the tape cassette 2, X-Y plotter 40, tape punch 41 and digital interface circuit 42 is controlled by a peripheral control circuit 43 through a pair of control lines 44 and 45. The peripheral control circuit 43 connects through a lead 46 to the least significant digit output of the instruction register 23.
Referring to FIG. 4, the basic system software components of the programmer comprise routines which are stored in the memory 10. When an on-off switch 50 is turned on a start-up routine 51 is executed to initialize certain variables and pointers and after its completion, control is passed to a keyboard monitor routine 52. The keyboard monitor routine calls for input data from the keyboard and display routine 53, and when such data is received, it searches mnemonic tables stored in the memory 10 to ascertain the identity of the statement or system command. When the input data is identified the system jumps to the appropriate routine. A separate syntax analysis routine is provided for each possible statement and an execution routine is provided for each possible system command. These are indicated collectively by the block 54.
Several of the statement and command execution routines require evaluation of arithmetic functions and expressions which is performed by formula evaluation routines 55. Other statements and commands require input from or output to one or more of the I/O devices, and accordingly, driver routines 56-59 for the respective I/O devices 42, 2, 41 and 40 are provided. When a statement or system command has been executed, control is returned to the keyboard monitor routine 52.
The keyboard and display routine 53 periodically refreshes the display 3 through a display driver routine 60 and receives data from an interrupt routine 61. When a key on the keyboard 1 is depressed the interrupt control circuit 31 (FIG. 3) causes the system to jump to the interrupt routine 61. The interrupt routine 61 saves the keycode in a specified memory location and control is returned to the interrupted routine. The keyboard and display routine 53 reads this memory location and decides what operations need to be performed in response to that particular keycode. For a more detailed explanation of the basic system software, reference is made to the above-cited U.S. Pat. No. 4,012,725.
To better understand the functions performed by the programmer unit of the present invention a description of the part program development procedure will first be made. This description will be made with reference to the example part in FIGS. 14 and 16 and the composite chart in FIG. 15.
As shown in FIG. 14, a typical part drawing is comprised of a series of straight lines and circular arcs which are concatenated to form its outline. If the starting point and end point of each straight line segment were given on the drawing along with the starting and end points of the circular arcs and their centers, the development of a part program for reproducing the illustrated part on a numerically controlled machine would be trivial. Such is never the case however. Instead, the geometry of the part is usually defined by the center points of circular arcs and their radii. The intersections of the circular arcs with each other and with straight line segments are usually not known.
It is a discovery of the present invention that a vast majority of parts made on numerically controlled machines can be constructed by concatenating selected geometric shapes. The number of different geometric shapes needed to form the vast majority of parts are relatively few in number and these can be defined by the geometric dimensions which are commonly available on part drawings. These selected geometric shapes are referred to herein as "composites," and as shown in FIG. 15, the preferred embodiment of the invention described employs a set of five separate composites which are available to describe a part. These are identified hereinafter as composites I-V.
It should be apparent to those skilled in the art that the number of composites and their specific features may be varied. For example, if a class of geometric shapes is encountered quite often in certain applications, then additional composites may be added which will facilitate the development of part programs to reproduce such shapes. On the other hand, some of the composites shown in FIG. 15 may be eliminated. For example, composite No. I can be viewed as a special case of composite No. II in which the radii of the circles at each end are both zero. The choice and number of composites employed involves a trade-off between the convenience to the operator which additional composites provide versus the cost of adding the composite to the programmer unit.
Referring particularly to FIG. 16, the method implemented by the programmer unit of the present invention begins at a starting point 65 on the shape to be programmed and travels in the direction of the arrow 66 around its perimeter. The shape of the part as one travels around its perimeter may be described by a sequence of composites selected from the available set I-V. Thus the illustrated part may be described by the successive line segments 67-72, the last of which ends at the starting point 65. The successive line segments are formed from selected ones of the set of composites I-V as follows:
Each composite No. I-V relates to a class of specific, but dimensionless geometric shapes. Having established that a segment of a part may be generally described by a particular composite, the task of specifically defining the line segment is reduced to the task of merely adding dimensions. This task is simplified because there are but a few dimensions, or parameters, needed to specify a shape from each of the composites I-V and all of these dimensions are readily available on typical engineering drawings. The set of parameters associated with each of the composites I-V are indicated in the following Table A.
The programmer unit asks the operator to select a composite number by displaying a message on the alpha-numeric display 3. The appropriate composite definition means is thus selected and the programmer unit sequentially displays messages which request the operator to input the dimensions for the selected composite. The above parameters are thus entered by the operator through the keyboard 1 on the programmer unit and they are stored in an (L) array 75 within the memory 10 along with their associated composite number. The elements of the (L) array 75 are identified by the row (R) and column (C) in which they are located. The contents of the (L) array are indicated in FIG. 17 and it can be seen that each row (R) stores a composite number and its associated parameters. It should be apparent that not all elements in a given row (R) will be meaningful since the parameters associated with each type of composite I-V are different. For example, when composite No. I is specified in column 1, only X and Y dimensions in columns 2 and 3 are stored. The remaining elements in that row remain blank and serve no purpose. The contents of the (L) array 75 after the part shown in FIG. 14 has been completely described is illustrated in FIG. 17. A "zero" in column 1 of the (L) array 75 indicates the end of the part description.
The programmer unit of the present invention interacts with the operator to obtain the identity of the successive composites which form the part, and with respect to each selected composite, to obtain the dimensions which define the geometric shape. The sequence of events is controlled by a main routine which is stored in the memory 10 of the programmer. A flow chart of the main routine is shown in FIGS. 5-13 and the operation of the programmer will now be described with respect to this flow chart.
Referring particularly to FIG. 5, the main routine is entered at the point 77 and a first instruction indicated by a process block 78 outputs data to the alpha-numeric display 3 which asks the operator whether an existing part program data file is to be employed or whether a new part program data file is to be developed. An instruction indicated by a process block 79 inputs the operator's response (Y or N) and a branch is established by a decision block 80. If an existing part program file is desired, an instruction indicated by a process block 81 couples data to the alpha-numeric display 3 which asks the operator to identify the desired file number. Responsive data is then coupled from the keyboard 1 as indicated by the process block 82 and the indicated part program file is then read in from the tape cassette 2 and stored in the (L) array 75 of the memory 10 as indicated by the process block 83.
If a new part program file is to be developed, a branch does not occur at the decision block 80. Instead, preparation for the development of a part program data file in the (L) array 75 is begun by setting a row pointer (R) to "one" as indicated by a process block 84. As indicated by a process block 85 the operator is then asked to enter the identity of the first composite on the part to be described. The response from the operator is inputted from the keyboard 1 as indicated by the process block 86 and a column pointer (C) for the (L) array 75 is then set to "one" as indicated by the process block 87. The identity of the selected composite number is then determined by a set of instructions which are indicated by the successive decision blocks 88-92. Having identified the selected composite (I-V) a corresponding composite routine indicated by the respective process blocks 93-97 is executed. As will be described in more detail hereinafter, each composite routine displays a succession of messages to the operator which asks him to enter the dimensions needed to define the selected composite. The data thus entered is stored in the line of the (L) array 75 indicated by the row pointer (R) at the position indicated by the column pointer (C). After the selected composite has been completely described, the system loops back to an instruction indicated by the process block 98 which is executed to increment the row pointer (R). The system then repeats the process of identifying the next composite needed to define the boundary of the part. The system remains in this loop until the operator enters a "zero" for the composite number. As will be described in more detail hereinafter the functions performed in this loop which is defined in FIGS. 5 and 6 by the dashed line 175 are accomplished by a single routine which is identified as the "combined composite routine."
The (L) array 75 is thus filled with data which defines the part as a series of composite numbers and associated dimensions. After the last composite has been identified the operator enters the number "zero" in response to the question posed by the process block 85. The loop is then exited to a process block 99 shown in FIG. 6.
Sets of instructions indicated by the process block 99 and process blocks 100 and 101 are then executed to output the contents of the (L) array to the printer along with an X and Y parity indication. If the part has been properly described by the operator and the descriptive data properly stored in the (L) array 75, the starting point and end point in that description will have the same X and Y dimensions. If a discrepancy occurs it will be indicated as an X or Y parity which may be examined along with the contents of the (L) array 75 to determine where the error was made.
Referring to FIG. 7, the programmer unit includes an instruction indicated by a process block 102 which asks the operator through the display 3 to indicate whether or not the contents of the (L) array 75 are correct. The response is inputted from the keyboard 1 as indicated by a process block 103 and it is then evaluated by instructions indicated by decision block 104.
If corrections are to be made to the contents of the (L) array 75, a loop indicated by an arrow 105 is entered and the operator is asked which row (i.e. composite) is to be changed. The appropriate composite is selected by the operator and the corresponding composite routine is executed to reenter all dimensions in the selected row of the (L) array 75.
If no corrections are to be made, or if no further corrections are to be made, the system branches in the opposite direction at the decision block 104. As indicated by process blocks 106 and 107, instructions are then executed which ask the operator through the display 3 whether or not the contents of the (L) array 75 are to be reprinted. If so, a loop indicated by an arrow 108 is established by a decision block 109. After the part has been completely and accurately described by the contents of the (L) array 75, the main routine proceeds to the process block 110 in FIG. 8.
Referring particularly to FIG. 8, an instruction indicated by the process block 110 outputs data to the display 3 which asks the operator whether or not the developed data file stored in the (L) array 75 is to be stored in the tape cassette 2. The operator's response is inputted from the keyboard 1 as indicated by process block 111 and if the file is to be stored, the system branches at a decision block 112 to a process block 113. The instruction indicated by the process block 113 outputs data to the alpha-numeric display 3 which asks the operator to enter the number of the data file stored in the (L) array 75. The response is received from the keyboard 1 as indicated by a process block 114 and the indicated file number along with the contents of the (L) array 75 are then outputted to the tape cassette 2 by means of instructions indicated by a process block 115. This completes the generation of a geometric definition of the part according to the present invention and subsequent steps in the procedure are employed to convert this definition into a part program of standard form which may be employed on a commercially numerically controlled machine to reproduce the part.
The definition of the part stored in the (L) array 75 is a very concise definition in that it does not require a great deal of memory space. It is also a convenient form in which to store the definition of a part since it can be easily altered when changes are to be made in the part. Before the part can be reproduced on a commercially available numerically controlled machine, however, the form of the definition must be converted to data which is understood by such a machine. In other words, it must be converted into a standard format and code which defines the part as a series of moves to be made by the machine.
Referring to FIGS. 2 and 18, an (O) array 116 is stored in the main memory 10. The (O) array 116 includes a plurality of rows, each of which includes five locations for storing data that defines a move to be made by the machine. Each location in a row is identified by a column number 1-5. The element in column 1 of each row in the (O) array 116 is a move code which is referred to in the art as a "G" code. The "G" codes are indicated in TABLE B.
The elements in columns 2 and 3 of each row in the (O) array are the respective X and Y coordinate values of the end point of the indicated move. The elements in columns 4 and 5 in each row are the respective X and Y coordinate values of the center point about which a circular arc is to be cut. By convention the latter two elements are referred to as the "I" and "J" values. It is a function of the invented programmer unit to convert the definition of the part which is stored in its (L) array 75 into a series of moves which are stored in successive rows of the (O) array 116.
Referring again to the flow chart of the executive routine in FIG. 8, the first step in this conversion process is to request from the operator the absolute coordinate values of the starting point of the machining process. This is accomplished by a set of instructions indicated by process blocks 117 and 118. The response from the operator is stored in the (O) array 116 as the endpoints of a first straight line move by a set of instructions indicated by process blocks 119 and 120.
As indicated by a process block 121, the operator is next asked to indicate whether the part is described in the (L) array 75 as a series of composites which travel clockwise or counterclockwise around the perimeter of the part. The response is received from the keyboard 1 and stored in the memory 10 for later use as indicated by a process block 122. An (L) array row pointer (R) and column pointer (C) are then set to "one" as indicated by process block 123 and a loop indicated by dashed line 176 is then entered in which the data on each row of the (L) array 75 is converted to a series of moves which are stored on successive rows of the (O) array 116.
Data is first read from the indicated row in the (L) array 75 into a temporary storage area as indicated by a process block 124. The composite number associated with this row is then determined by a set of instructions indicated by respective decision blocks 125-129 and the system branches either to one of the corresponding composite processor routines 130-134, or it branches out of the loop to a process block 135 shown in FIG. 10. As will be described in more detail hereinafter, the composite processor routines 130-134 are each associated with a respective composite I-V and each converts the data on the indicated row of the (L) array 75 to a series of moves which are stored on successive lines in the (O) array 116. After the appropriate composite processor routine 130-134 is executed, the system returns to a process block 136 shown in FIG. 9 which increments the row pointer (R) and loops back to the process block 124. The process of converting each row of the (L) array 75 to data stored in the (O) array 116 continues until the composite number "zero" is detected at the decision block 125. The end of the part description is thus indicated and the system branches to the process block 135 in FIG. 10. As will be described in more detail hereinafter, the functions performed within the loop defined by dashed line 176 are governed by a program identified as the "composite processor routine."
An instruction indicated by the process block 135 outputs data to the alpha-numeric display 3 which asks the operator whether or not the part defined by the data in the (O) array 116 is to be generated on the X-Y plotter 40. The operator's response is inputted from the keyboard 1 by an instruction indicated by process block 136. The response is analyzed by instructions indicated by a decision block 137. If a plot is to be made, the system branches to an instruction indicated by process block 138 which asks the operator to indicate the scale at which the part is to be reproduced. The responsive data is inputted from the keyboard by an instruction indicated by process block 139 and then the operator is asked to indicate the cutter offset by an instruction indicated by process block 140. The cutter offset characters are inputted from the keyboard 1 by instructions indicated by process block 141 and then instructions indicated by process block 142 are executed to set an (O) array row counter (RO) and column counter (CO) to "one."
Referring to FIG. 11, a loop indicated by an arrow 143 is then entered in which the move indicated by each line of the (O) array 116 is translated to output data for the X-Y plotter 40 to control its X and Y servomechanisms. More specifically, the data on the indicated row of the (O) array 116 is read out and stored in temporary registers by instructions indicated by a process block 144. The move code "G" is then analyzed by instructions indicated by process block 145. The move code "zero" indicates the part has been completed and in response to its presence, the system branches back to the process block 135 in FIG. 10. Otherwise, either a straight line or circular move is to be made.
As indicated by process block 146 calculations are then made to alter the contents of the (O) array 116 to reflect the indicated cutter offset. This is accomplished by a cutter offset calculation routine which examines the move codes and elements on three successive lines of the (O) array 116 to calculate the offset elements of one line. Such routines are well known in the art and the routine 146 employed herein is listed in Appendix A. After calculating the offset values of the move, a circular or linear interpolation routine indicated by process block 147 is executed to generate command data to the plotter 40. The interpolation routine 147 is listed in Appendix B. At the end of the indicated move an instruction indicated by process block 151 is executed to increment the row counter (R) and the system loops back to the process block 144 to analyze and execute the next move stored in the (O) array 116.
Referring again to FIG. 10, if no further plots are to be made, the system does not branch at the decision block 137 and an instruction indicated by process block 152 is executed to ask the operator to input cutter offset. An instruction indicated by process block 153 inputs the cutter offset characters from the keyboard 1 and the operator is then asked to input any start-up machine codes as indicated by process block 154.
Referring to FIG. 12, these start-up machine codes are inputted from the keyboard 1 and stored in temporary registers as indicated by process block 155. The (O) array row counter (RO) and column counter (CO) are set to "one" by instructions indicated as process block 156. A loop indicated by an arrow 157 is then entered in which the elements of the (O) array 116 are changed to reflect cutter offset. The data on the indicated line of the (O) array 116 is written into a temporary register as indicated by process block 158 and then the move code is identified as indicated by decision block 159. As indicated by process block 162, if the move code is a "nine", a "two" or a "three", the CUTTER offset calculation routine described above and listed in Appendix A is called up and executed. The resulting altered elements of the move command along with the move code are coupled to the tape punch 41 by a set of instructions indicated by process block 163. The (O) array row pointer (RO) is then incremented as indicated by process block 164 and the system loops back to read out the next line of the (O) array 116. When the move code "zero" is detected, the entire part program has been developed and the system branches at the decision block 159 to FIG. 13.
The last step in the development of the part program stored in punched tape is to enter any machine codes which are to follow the cutting of the part. More specifically, an instruction indicated by process block 165 outputs data to the alpha-numeric display 3 which instructs the operator that machine codes may be entered. Sets of instructions indicated by process blocks 166 and 167 input any such machine codes from the keyboard 1 and output the same to the tape punch 41. Because it is often desirable to repeat the part program, an instruction indicated by process block 168 is executed to output data to the alpha-numeric display 3 which asks whether or not the part program stored in the (O) array 116 is to be repeated. The responsive characters are inputted from the keyboard 1 as indicated by process block 169 and analyzed by instructions indicated by decision block 170. If the part program is not to be repeated the system loops back to process block 165, otherwise, the system loops back to the process block 152 in FIG. 10 to repeat the conversion of (O) array data to final form. For example, a tool change might be made, a new cutter offset entered and the part program repeated to perform a finishing cut.
As indicated above, the programmer unit interacts with the operator to load the data into the (L) array 75 which defines the part in terms of a series of composites. As described with reference to FIGS. 5 and 6, the operator is asked to identify the composite number and the programmer unit them calls up the appropriate composite routine 93-97. The composite routines 93-97 include instructions which interrogate the operator through the alpha-numeric display 3. These interrogations are requests for the operator to enter through the keyboard 1 geometric data which specifically defines the selected composite. The composite routines 93-97 input this geometric data from the keyboard 1 and store it along with the associated composite number in the (L) array 75.
Although in concept there is a separate composite routine 93-97 for each of the selectable composites I-V, in practice a single program with appropriately placed jump instructions is employed because it is shorter and hence requires less memory space. This routine which was identified above as the "combined composite routine," is as follows: ##SPC1##
As described above with reference to the flow chart of FIGS. 9 and 10, the composite numbers and associated geometric data stored in each row of the (L) array 75 is converted by composite processor routines into a series of moves which are stored on successive lines of the (O) array 116. Although in concept there is a separate composite processor routine for each of the selectable composites I-V, in practice a single program identified above as the "COMPOSITE PROCESSOR ROUTINE" is employed to process all of the composite numbers. This routine in turn employs a number of subroutines which are also used for all of the composites. The COMPOSITE PROCESSOR ROUTINE along with the "SUBROUTINE FOR CALCULATING TANGENCY POINTS (A,B,C,D) AND CENTER OF CONNECTING CIRCLE (I,J)", "(O) ARRAY MOVE STORAGE SUBROUTINE," and the "SUBROUTINE FOR CALCULATING 0, 90, 180 AND 270 DEGREE COORDINATES BETWEEN ESTABLISHED TANGENCY POINTS ON A CIRCLE" which it employs are as follows: ##SPC2## ##SPC3##
From the above listings it should be apparent that the five composites I-V are broken down into eleven subtypes. These subtypes are illustrated generally in FIG. 19 and it can be seen that composite II is broken into two subtypes (1 and 2), composite III is broken into two subtypes (3 and 4), composite IV is broken into four subtypes (5-8), and composite V is broken into two subtypes (10 and 11). In most instances the subtypes are distinguished by the direction of travel (cw or ccw) around the respective circles. The reason for establishing the subtypes is that the formulas for calculating the tangency points at which the connecting line intersects the circles differs for each.
In order to convert the composite numbers and associated geometric data into a series of moves, the coordinate values of the tangency points at which the connecting line intersects the circles must be calculated for all but composite I. In the case of composites III and IV the coordinate values of the center of the connecting circle must also be calculated. These calculations are made by the above listed SUBROUTINE FOR CALCULATING TANGENCY POINTS (A, B, C, D) AND CENTER OF CONNECTING CIRCLE (I, J). To better understand how these calculations are performed reference is made to FIGS. 20a and 20b in which the variables employed in this subroutine are shown as they apply to a subtype 1 and a subtype 2 geometry. The dimensions for G and H as well as the radii N, F and F1 are given (i.e. they are elements of the (L) array).
A programmer unit has been described in which a processor is programmed to interact through an alpha-numeric display and a keyboard with an operator to develop a part program for a numerically controlled machine. It should be apparent to those skilled in the art, however, that a hardwired processor may also be employed to perform the same functions. It should also be apparent that portions or all of the components of the invented programmer unit may be included as an integral part of a numerical control system. In such case the processor, (L) array and (O) array storage means of the present invention may be an integral part of the numerical control processor and memory or they may be a separate microprocessor and associated memory device or devices. These and other variations may be made without deviating from the spirit of the invention and reference is, therefore, made to the following claims for a definition of the invention. ##SPC4##
Claims
I claim:
1. A programmer unit for developing a part program for a numerical control system, the combination comprising: a memory for storing a plurality of multi-bit words, including a portion designated as an (L) array and a portion designated as an (O) array; data bus means coupled to said memory means for writing data into selected locations in said memory and for reading data out of selected locations in said memory; alpha-numeric display means coupled to the data bus means, said display means being operable in response to selected data on said data bus means to display messages for an operator; keyboard means coupled to the data bus means, said keyboard means being operable in response to manual operations by the operator to generate data to the data bus means; and processor means coupled to said data bus means and being operable to read program instructions out of said memory means and in response thereto perform a plurality of functions which comprise: coupling data to said display means which generates a message prompting the operator to select and enter one of a plurality of selectable composite numbers; coupling data from the keyboard means which represents a selected composite number to a location in the (L) array portion of said memory; coupling data to said display means which generates a message prompting the operator to enter a set of geometric data which is associated with the previously selected composite number; coupling data from the keyboard means which represents the set of geometric data to locations in the (L) array portion of said memory; and sequentially reading the composite numbers and associated geometric data out of said (L) array portion of said memory and converting each into a set of move codes and associated geometric data which are stored in the (O) array portion of said memory.
2. The programmer unit as recited in claim 1 in which a part program storage means is coupled to said data bus means and said processor means functions to sequentially couple the contents of the (O) array portion of said memory to said part program storage means.
3. The programmer unit as recited in claim 1 in which a plotter is coupled to said data bus means and said processor means functions to sequentially read out the move codes and associated geometric data from the (O) array portion of said memory and in response thereto generate command data which is coupled to the plotter to control its servomechanisms through a series of moves.
4. A programmer unit for developing a part program for a numerical control system, the combination comprising: a memory for storing a plurality of multi-bit words, including a portion designated as an (L) array which stores a plurality of composite numbers and associated geometric data and a portion designated as an (O) array which stores a plurality of move codes and associated geometric data; data bus means coupled to said memory means for writing data into selected locations in said memory and for reading data out of selected locations in said memory; alpha-numeric display means coupled to the data bus means, said display means being operable in response to selected data on said data bus means to display messages for an operator; keyboard means connected to the data bus means, said keyboard means being operable in response to manual operations by the operator to generate data to the data bus means; and processor means coupled to said data bus means and being operable to read program instructions out of said memory means and in response thereto perform a plurality of functions which comprise: coupling data from the keyboard means which represents a selected composite number and storing it in a location in the (L) array portion of said memory; coupling data from the keyboard means which represents a set of geometric data associated with said selected composite number and storing the same in locations in the (L) array portion of said memory; and sequentially reading the composite numbers and their associated geometric data out of said (L) array portion of said memory and converting each into a set of move codes and associated geometric data which are stored in the (O) array portion of said memory.
5. The programmer unit as recited in claim 4 in which each of said composite numbers is associated with a specific class of geometric shapes and the geometric data associated with each composite number defines a specific geometric shape within the class of possible geometric shapes.
6. The programmer unit as recited in claim 5 in which said move codes include: a straight line move code in which the geometric data associated therewith defines the location of the end point of a straight line move; and a circular move code in which the geometric data associated therewith defines the location of the end point of a circular move and the location of the center of the circle about which the circular move is to be made.
7. The programmer unit as recited in claim 6 in which said processor means functions in response to at least one of said selectable composite numbers read from said (L) array to convert it and its associated geometric data into a circular move code and associated geometric data, a straight line move code and associated geometric data and a second circular move code and associated geometric data.
8. The programmer unit as recited in claim 7 in which said processor means functions in response to at least one other of said selectable composite numbers read from said (L) array to convert it and its associated geometric data into a series of three circular move codes and their associated geometric data.
9. The programmer unit as recited in claim 8 in which said processor means functions in response to at least one other of said selectable composite numbers read from said (L) array to convert it and its associated geometric data into a series of two circular move codes and their associated geometric data.
10. The programmer unit as recited in claim 6 which includes means for coupling said data bus means to the servomechanisms on a machine and in which said processor means functions to read out straight line move codes and their associated geometric data stored in the (O) array portion of said memory, generate command data therefrom and output it to said servomechanisms; and said processor means also functions to read out circular move codes and their associated geometric data stored in the (O) array portion of said memory, generate command data therefrom and output it to said servomechanisms.
11. A programmer unit for developing a part program for a numerical control system, the combination comprising: a memory for storing a plurality of multi-bit words, including a portion designated as an (L) array and a portion designated as an (O) array; data bus means coupled to said memory means for writing data into selected locations in said memory and for reading data out of selected locations in said memory; I/o control unit means connected to said data bus means for receiving data therefrom and coupling data thereto, said I/O control unit means being connectable to a plurality of I/O devices through I/O data bus means; alpha-numeric display means connected to the I/O control unit through said I/O data bus means, said display means being operable in response to selected data on said I/O data bus means to display messages for an operator; keyboard means connected to the I/O control unit through said I/O data bus means, said keyboard means being operable in response to manual operations by the operator to generate data to the I/O control unit means; and processor means coupled to said data bus means and being operable to read program instructions out of said memory means and in response thereto perform a plurality of functions which comprise: coupling data to said display means which generates a message prompting the operator to select and enter one of a plurality of selectable composite numbers; coupling data from the keyboard means which represents a selected composite number to a location in the (L) array portion of said memory; coupling data to said display means which generates a message prompting the operator to enter a set of geometric data which is associated with the previously selected composite number; coupling data from the keyboard means which represents the set of geometric data to locations in the (L) array portion of said memory; sequentially reading the composite numbers and associated geometric data out of said (L) array portion of said memory and converting each into a set of move codes and associated geometric data which are stored in the (O) array portion of said memory.
12. The programmer unit as recited in claim 11 in which a part program storage means is coupled to said I/O control unit means and said processor means functions to sequentially couple the contents of the (O) array portion of said memory to said part program storage means.
13. The programmer unit as recited in claim 11 in which a plotter is coupled to said I/O control unit means and said processor means functions to sequentially read out the move codes and associated geometric data from the (O) array portion of said memory and in response thereto generate command data which is coupled to the plotter to control its servomechanisms through a series of moves.