Comprehensive design and maintenance environment for test program sets4894829Abstract A source computer program for generating object computer programs, referred to as test program sets, for use in operating a particular automatic test equipment unit (ATE). The ATE is useable for testing a variety of electrical equipment type units each of which is referred to with relation to said ATE as a unit under test (UUT). The source program involves (1) having a test engineer, who designs the individual tests, enter the tests graphically at a work station, (2) verifying the match between the entered test design, the ATE capabilities and the UUT description, and (3) having an automatic way of generating a test program for a test program set from the stored semantic content of the graphically entered test design. Claims It is claimed: Description The invention relates to a comprehensive design and maintenance environment for test program sets.
______________________________________
Direct Current Power
Spectrum Analyzer
Supplies
Alternating Current Power
Waveform Digitizer
Supplies
Radio Frequency Power
Peak and Average Power Meters
Sources
Electro Optical Sources
Counter Timers
Digital Pattern Generators
Oscilloscopes
Volt Meter Communication busses
Ohm Meter Impedances
Current Meter
______________________________________
The signals provided by the stimuli are routed within the ATE and delivered to an ITA which is represented by the interface unit 12 of FIG. 1. The ITA 12 is where the UUT 10 is mated to the ATE 14. The routing of signals within the ATE is controlled by various switches and relays by which it is possible to deliver signals to the UUT to stimulate a particular segment of its circuitry. The resultant response from the UUT is then fed back to the ATE where it is routed to the appropriate measurement device. In tests in which the measured value is compared to known limits, the result of the comparison determines the next test with regard to the stimulus to be applied, the measured response thereto and the determination of the validity of the response. The ATE computer communicates with and controls the ATE resources over communication busses using standard protocols. A user of an ATE unit programs the computer thereof to control the sequence of tests by encoding the testing logic in a high level language such as ATLAS. The user can specify the parameters of the signals delivered within the capabilities of the ATE resources. Similarly, the ATE resources can measure specific parameters of the signals received from the UUT and communicate the result to the application software wherein the testing logic executes the next series of steps. In this manner the faults within the UUT may be determined and the operator is directed to take appropriate corrective action. UUT's typically have hundreds of input and output pins as indicated by the block diagram of the RF pulsed amplifier of FIG. 2. Manufacturers of UUTs issue specifications and performance tests for these products which are the basis for tests which involve applying known signals to input pins and sensing the responses thereto of an UUT at the output pins thereof. A test engineer will study the specifications and documentation of a UUT and design a series of perhaps hundreds of performance and diagnostic tests to be performed on a UUT by the ATE. Each test is designed to detect malfunctioning in a subcircuit which is of interest and a test program will normally include several hundreds of such tests. In prior installations, it is the practice to utilize the services of a computer programmer to prepare a program based on the designed tests for operating the ATE. As an alternative, if the engineer who designed the tests might also have some programming expertise, he might then also prepare the test program. In the invention herein a computer program is provided for the test engineer which has two basic functions. These functions are (1) to provide means for aiding the designing and maintaining of the tests, and (2) to provide means for automatically encoding the design results to form a test program set which comprises the test program, the ITA and associated documentation which will be used to operate the ATE 14 of FIG. 1. The program embodying the invention herein has the name CoMETS which is an acronym for a Comprehensive design and Maintenance Environment for Test Program Sets. CoMETS is coded using the object-oriented programming language known as Smalltalk-80.TM.. CoMETS is useable on a continuing basis for producing programs to control ATEs in performing tests on a variety of UUTs. In this instance, the test programs so produced are encoded with the language ATLAS. The work station for the design engineer is a computer upon which the CoMETS environment is run. FIG. 4 is a block diagram illustrating how a test program set, which includes a test program, is developed. The personal workstation interface 24 is interactive and provides the means for a test engineer 22 to enter station information 32, UUT description 34 and test design data into the test program development system 20. The test program development system 20 comprises a personal work station interface 24, a test generating system 26, and a data base 27 that is used for generating a test program set 81 which includes test program 28 having ATLAS source code, an ITA electrical design 80, and other associated reports 29. The test program 28 is used for operating the ATE 14 which will perform the required tests on the particular UUT 10 for which the test program set has been designed. The test generating system 26 comprises a processor and checker 30, an ATLAS converter 37, the internal representation 36 of the data (station information 32, UUT description 34 and test designs comprising the TPS under development), an ATLAS generator 72, and a report generator 73. The ATLAS converter 37 provides for the direct entry of ATLAS code to the work station interface 24. The ATLAS converter 37 converts the ATLAS code into the same internal representation in section 36 as that done by entry in the graphical mode. The ATLAS generator 72 generates the test program 28 having ATLAS source code by converting the test design for the UUT 10 stored in the data base 27. The report generator 73 will generate additional reports that are required. These reports include an updated test requirements document, a test strategy report, etc. In addition, the report generator also produces the electrical design for the ITA 12 (FIG. 1). The test engineer 22 is shown as the recipient of the following data: station information 32, a UUT description 34, a test requirements document (TRD), UUT drawings, and technical orders relating to the ATE 14 and the UUT 10. The test engineer synthesizes this information and designs a test program. The test engineer 22 first inputs the station information 32 into the test program development system 20 via the workstation interface 24. This information is stored in an internal representation 36 and also stored in the data base 27. The station information 32 is entered once for each ATE station. The test engineer 22 then enters the UUT description 34 into the test program development system 20 via the workstation interface 24. This information is stored in an internal representation 36 and also stored in the data base 27. The UUT description is entered once for each UUT 10 being tested. The test engineer 22 next enters the test designs from the data contained in the test requirements document, UUT drawings and technical orders. The test design data is entered graphically or textually into the test program development system 20 via the workstation interface 24. As test designs are entered, the processor and checker 30 checks that the test design is consistent with the UUT 10 and ATE 14 by validating entries against the station information 32 and UUT description 34. In designing the test program, the CoMETS tool set provides significant productivity improvement. The TPS engineer is relieved of the burden of expressing a design approach in the test language, namely ATLAS. With CoMETS, the designed focuses attention on the design of the test and its position in the test flow, and not on the details of the implementation language. By capturing the test strategy in a logically understandable flow diagram, the TPS engineer can focus attention on improving the quality and integrity of the design. The graphical representation of the test strategy is captured in an object oriented data base 27. As the design data is input, the data's validity, in terms of being able to produce a workable test program is checked against the station information 32 and UUT description 34. The station information 32 and UUT description 34 are stored in the object-oriented data base 27. The station information 32 defines the ATE environment such as the available resources and their functional parameters, routing of the signals within the ATE, mating data between the ATE and ITA, etc. The UUT description 34 defines the UUT relative to the input/output pins, signal types and their limiting parameters, etc. Mapping the design data against these data (ATE 32 and UUT 34 information) is key to ensuring the is resulting test program will execute on the ATE. The UUT 10 illustrated in FIG. 2 has the specifications listed in Table 1 of FIG. 3. The test objectives to checkout the circuitry of the UUT are set forth in Table 2 of FIG. 3. In designing the tests for the UUT, the test engineer will have available to him interactive screens 38 and 40, illustrated in FIGS. 5 and 6, of the work station interface 24 which are supported by CoMETS. The CoMETS system may be used in many different scenarios. For example, at the start of a design session the screen 38 is brought up and the "Title" box 42 is activated to provide an identifying name for the test program 28. The test program might be given a suggestive name tying it to the UUT to be tested and such program will eventually appear between the main "Start" and "Stop" boxes 44 and 45 on the left side of the screen 38. Although only three tests 46, 47 and 48 are indicated for the purpose of illustration, a test program for a UUT will typically have hundreds of such tests. After the test program is named, the screen 40 of FIG. 6 is brought up. Four stages a, b, c and d of the screen 40 are illustrated. CoMETS lets the TPS engineer set up individual tests with great ease as shown in FIG. 6. In the CoMETS tool set, the focus of the interface is shifted from a tool meant to help a user input correct statements to a tool meant to let the user concentrate on the design of a test. The techniques herein allow a user to use icons representing physical entities with which the engineer is comfortable. The metaphor is that of an engineer at a work bench. The user graphically connects the symbols to represent a test set-up. Collections of these low level building blocks (icons) are used to represent complete tests. Next the user links the tests to establish a logical flow based on the test strategy. The individual tests such as tests 46 to 48 are constructed by the use of graphical symbols and a combination of pull-down and pop-up menus. The concept helps the TPS engineer generate syntactically and semantically correct test programs in ATLAS by use of a friendly, well understood interface. The engineer constructs the test program test by test. However, it is not necessary for the user to have a complete understanding of the ATLAS language rules. The user can control the flow of the test program; add and delete tests; alter the contents of specific tests; create libraries of tests; browse through tests; incorporate tests from the library; etc. FIG. 6 illustrates the tests in testing the radio frequency hardware illustrated in FIGS. 2 and 3. Three separate setups were performed to generate a valid test. This exercise illustrated that the user did not need a complete understnding of ATLAS to generate valid tests. FIG. 2 illustrates the block diagram of a test circuit frequently encountered in the radio frequency (RF) domain. Table 1 of FIG. 3 contains its specifications and Table 2 of FIG. 3 lists the test objectives that need to be implemented to satisfactorily test the circuitry of FIG. 2. The left side of the screen 38 of FIG. 5 is the representation of the three performance tests 46, 47 and 48 (of which as many as several hundreds are required) designed to check out the circuitry of FIG. 2. Referring to the screen representations of FIG. 6 it is assumed that a test requirement document (TRD) describing the tests which are to be run on the UUT has been assembled. The TRD has schematic diagrams and descriptions of tests which are to be run. The test engineer brings up the screen 40 and he activates the test box 50. This brings up a menu for adding or deleting a test. He chooses to add a test and, after activating the title box 52, names the test "Power Test". The "Power Test" is added in the same manner as the next test, "RF mod>10%", and will not be separately described. After the "Power Test" is added, the test engineer again activates the test box 50 to facilitate adding the next test. He activates the title box 52 and names the test "RF mod>10%". For this test he activates the SOURCE box 54 and from the resulting menu (not shown) he chooses the "PULSED DC" signal source which appears in the form of an icon 56 in screen 40a. The format of screen 40 has a pin connection bar 58 formed by separate connection boxes representing input/output ports on the UUT. These boxes can be labled with UUT pin numbers to which chosen sources, sensors and loads can by symbolically connected. In this case the source 56 is connected as to two boxes of the bar 58 which are given the pin connection labels J4-9 and J4-10. Identification of the selected source and pin numbers are stored in the data base 27 of the TPS development environment unit 20 (CoMETS) shown in FIG. 4. The available parameters for the "PULSED DC" source 56 are accessed from a menu 60 (screen 40b) which is obtained by touching the icon 56. Means such as an "on screen" calculator (not shown) is provided to provide numerical values for the choosen parameters. The test engineer again activates the SOURCE box 54 and from the resulting menu (not shown) he chooses the "RF" signal source which appears in the form of an icon 62 in screen 40c. The source RF is connected to a box of the bar 58 and given the pin connection label J3. This data is likewise stored in the data base 27. Parameters or modifiers may be chosen for any source or sensor the same way as illustrated with respect to the source "PULSED DC". The test engineer next activates the SENSOR box 64 and from the resulting menu (not shown) he chooses the "RF Pulsed AC Power Peak" which appears in the form of an icon 66 in screen 40d. This sensor is connected to a box of the bar 58 and given the pin connection label J6. Parameters or modifiers for this sensor may be chosen if desired or necessary in the same manner as for the sources as referred to above. This data is likewise stored in the data base 27. This completes the entry of data for test 47 (FIG. 5) and the test engineer next proceeds to test 48. Data for all of the tests 46 to 48 is converted for storage and stored into an object-oriented data base 27. After the sources and sensors are selected for a test the "TEST" box 66 of screen 38 is activated and a menu (not shown) is produced which allows sensor choices for each test of GO/NO GO, High/Low and parameters such as = to, > or < than, etc. Activation of the "TPS" box 68 facilitates linking all the tests 46 to 48 into a complete "test program" as shown on the left side of the screen 38 of FIG. 5. This is at the "test interconnection level" (TIL). Finally, the "ATLAS" box 70 of screen 38 is activated and this causes the automatic generation of test program 28 using ATLAS code via the ATLAS GENERATOR 72 (FIG. 4) for the test program set 81 (FIG. 4) on the right side of the screen 38. Each time the test engineer places an icon representing a device on the screen, an object representing the device is created in the internal representation. When the test engineer specifies attributes of the device, such as signal level or frequency, that information is automatically added to the internal object that represents it. When the test engineer connects the icon to the pin connection bar 58 (FIG. 6) the connection between the object representing the bar 58 and the icon is stored in the internal representation. Similarly, when the test engineer starts designing a new test an object is created to represent the test in the internal representation, and when the engineer specifies flow-chart connections between tests, such as between "Power Test" 46 and "Fail Power Test" (FIG. 5), the connection between the objects representing the two tests is stored in the internal representation. Thus as the test engineer proceeds, a hierarchical structure is created step by step that faithfully reflects the structure of the TPS and the definitions of the individual tests that make it up, as completely as the test engineer has defined them at that point. Part of the definition of an object corresponding to a device or test is the method for generating the ATLAS code for itself, possibly including sending requests to the objects that represent its parts requesting them to generate the ATLAS for themselves. Thus the ATLAS for any part of a test program can be generated by sending a message to the object that represents it, requesting it to produce the code for itself. At generation time a message is sent to each test in the test program flow to generate ATLAS. Each object in the test has attributes associated with it that identifies what ATLAS will be generated. When a test gets a message to generate ATLAS, each object presents its piece of ATLAS to the ATLAS generator which combines that information with other pertinent information and produces ATLAS code for viewing on the display window. The overall design procedure may be summarized as follows: A. The user designs a test by using a graphical interface to choose resources, signal parameters, pin connections, etc. B. As the user builds the test, internally CoMETS collects the data objects that represent what he is designing. C. The user describes the flow of the test program by connecting tests in a specific order in a test program flow. D. The user may generate ATLAS for the test program flow (all tests included in the flow). E. When ATLAS is generated, each test is sent a request to generate ATLAS. F. When each test receives the message to generate, it sends a request to each of its data objects to generate ATLAS. G. Each data object has information stored with it that describes the ATLAS it should generate. This is the ATLAS that is generated. The overall effect is that after the test engineer enters the design of the tests graphically, the contents of the data base 27 are processed to produce a correct ATLAS program complete with the preamble and procedural statements.
|
Same subclass Same class Consider this |
||||||||||