JPH0330258B2 - - Google Patents

Description

[Detailed description of the invention]

TECHNICAL FIELD The present invention relates to methods for testing, testing, analyzing and diagnosing electronic networks or circuits and other devices. Two main methods are known in the art for testing electronic circuits, such as printed circuits and integrated circuits. 1st
This method is called a functional test, in which an appropriate signal is applied to the input terminal of a circuit and the voltage, current, or both at the output terminal is checked. The input signal can be varied and the corresponding variations in the output signal examined until the operation of the circuit is determined to be correct over the entire design range of the input signal. When a circuit responds correctly to all different test inputs, it can be assumed that all parts of the circuit operate correctly, but it is practically difficult to completely test a complex circuit in this manner. It is also known to test a circuit that is energized as if it were in normal operation by internal probing. It is also known to test electronic circuits by preparing a program which specifies a test sequence, for example to determine the voltage (associated with some data) at each successive point in the circuit. The program can be executed using a data processor, continuous voltage values, and possibly other electrical parameters provided as the test program progresses. Such a program can indicate faults that cause a particular voltage to be out of tolerance for a particular point being tested. The presence of a fault is confirmed by the computer and the program can direct a routine or method to locate the specific fault source. The routine has a series of instructions that position the probe to a particular set of points within the circuit. A machine in place of a hand-held probe that moves and engages various points in the circuit under test under automatic control of test equipment and that executes a test program at least in part automatically under the control of a computer program. It is also known to use targeted probes. The electronic circuit program testing methods described above are known in the art, and the formulation of test programs and specific test programs are not a feature of the present invention. Mechanical probe measurements are commonly used, but are slow and cumbersome for complex printed circuits and integrated circuits, and are not commonly used for modern functional testing. It is an object of the present invention to enable quick and easy post-assembly testing of complex printed circuits and integrated circuits;
An object of the present invention is to provide a method for testing an electronic circuit network that can normalize changes in the position of test points and minute changes in the angle of incidence of a beam. The invention is based on the use of an electronic probe that directs a deflectable, amplitude modulated, focused electron beam to a series of points on the surface of an electronic circuit, mediated by an operator, but usually a computer. A secondary electron current is generated from the electronic circuit by the action of the electron beam constituting the electron probe. This secondary electron stream is identified and only those electrons whose kinetic energy or direction of motion relative to the impact point, for example, is within a specific range according to specific identification criteria, pass through. The parameters are the surface potential, i.e. the voltage at the impact point where the electron beams that make up the electron probe collide (with respect to the data), the electric field near the impact point of the electron probe and near the surface of the electronic circuit under test, or the voltage at the point of impact of the electron probe. It can be an electromagnetic field near and near the surface of the electronic circuit under test. In the embodiments to be described later, an electron filter is a combination of a filter for secondary electrons and intermediate electrons, an electrode or surface that absorbs the residual secondary electrons after passing through the filter, and ancillary components that ultimately carry a signal representative of said parameter. The collector is designed to generate a signal representative of the surface potential at each test point and is called a voltage-sensing collector, but the disclosed device is designed to easily detect other parameters when desired. Can be changed. Thus, for example, when using a voltage-sensing collector, the electron beam constituting the voltage probe is directed to impinge on a predetermined connection point of the electronic circuit under test;
As long as the voltage at that location does not change independently of the electronic circuit, the voltage sensing collector can provide an output signal indicative of the voltage at that node. For any given combination of voltages or currents applied to the input terminals of an electronic circuit under test, the voltages at a given sequence of many connection points within the electronic circuit will change because the orientation of the electronic probe can be easily changed. Can be measured quickly. Alternatively, the electronic probe may be aimed precisely at a particular connection point while changing the input to the electronic circuit, or the test program may combine the testing of a large number of connection points with various changes to the input of the electronic circuit. can. Figure 1 shows a vacuum enclosure, a device that generates, focuses, and deflects an electron beam that constitutes an electron probe, and an electronic circuit that is a double-sided printed wiring board with the circuit components facing downward. A voltage sensing electron collector for detecting secondary electrons is shown.
The electronic probe is applied to the back side of the printed wiring board, and each test point, or connection point that holds the electronic probe stationary for surface potential measurements, is a conductive track at or near the solder mass at the terminal wire end of the component. The electron beam constituting the electron probe 1 is generated by an electron gun 2, but this electron gun is somewhat different from a normal electron gun used in a cathode ray tube in that the cathode is removable. The glass cone and screen of a typical cathode ray tube have been removed. The electron gun 2 is mounted as usual in a glass neck 3, the end of which is removably sealed to the narrow end of a metal (mild steel) cone 4. The focusing coil 5 and deflection coil 6 of the electron probe are mounted in a cross-suspension to a frame 7 extending concentrically from a flange 8 at the narrow end of the cone 4 for precise adjustment. A tube 9 extending from the back of the glass neck 3 and used for evacuating a normal cathode ray tube, and then permanently cut out, is left open and removably connected to a vacuum pump (not shown), and is connected to an electron gun. Keep the area under high vacuum. The glass tube 9 is connected to the electrical connection terminal 11 of the electron gun.
It is surrounded by a shield 10 made of a high magnetic permeability alloy. At the bottom of the cone 4 is a chamber mainly defined by a cylindrical frame 12, the inner end flange of which is sealed against the mouth of the cone 4, and at its base 13 a mouth connected to another vacuum pump. During testing, the electronic circuit area is kept in a high vacuum. The electronic circuit is mounted perpendicularly to the deflection direction of the electron beam constituting the probe 1 by a holding member, and the holding member has a support arm 14 for holding it on one side of the printed wiring board 25, and the printed wiring On the other side with the terminals of the electronic circuit on the plate there is a plate 16 with an edge connector 15 and on the inner surface of this connector 15 an O-ring 17 which engages with the flat outer surface of the side wall 18 of the test chamber. It is equipped with Therefore, printed wiring board 2
All of the input/output terminals of No. 5 are located outside the vacuum chamber, and their connections can be made in the air. The mounting plate 16 and the test circuit mounted thereon can be inserted horizontally into the test chamber. A pre-evacuated cartridge with multiple test circuits can be used if desired. Furthermore, new test circuits can be quickly converted to circuits that have been tested, and the test chamber can be evacuated quickly. High permeability alloy shield 19, 20, 21
surrounds the cone 4 and the test chamber. The voltage sensing electronic collector consists of a combination of components as described below. Closely parallel to the back side of the printed wiring board is a first grid 23 of electron transparent planes. It is defocused with respect to the primary electron beam consisting of the primary electron beam constituting the probe 1, and the pitch of this grid 23 and the width of each grid line are such that the electron beam constituting the probe is not focused on the printed wiring board 25. The selection is made so that changes in the primary electron beam when scanning the back surface 24 can be ignored. In the example, grid 2
3 is maintained at a positive voltage of 500V, and all low-energy secondary electrons emitted from the test point are transferred to the first
Accelerated towards the grid 23, most of these secondary electrons penetrate the first grid 23. Parallel to the first grid 23 and adjacent to its upper side is a second planar grid 26 which is also electron transparent. This grid is also defocused with respect to the primary electron beam and its pitch and the width of each grid line are similar to the first grid 23, so the second grid 26 is The first is to allow both parties to pass through these grids as much as possible.
It can be arranged relative to the grid 23. These secondary electrons penetrating the first grid 23 are
A retarding electric field is encountered between grid 23 and second grid 26. Usually the second grid is maintained at a voltage _V2 between 0 and -5V with respect to the potential of the test point. A potential difference may exist between the test point and the second grid 26. This can be corrected by appropriately setting the second grid voltage. If the test point is 0V, when the charge emitted from the test point is e and the mass of the electron is M, the velocity component perpendicular to the second grid is

[Formula] Secondary electrons larger than m/sec penetrate the second grid, and secondary electrons whose vertical velocity component is smaller than this cannot penetrate the second grid. For example, if the potential at the test point is -2.5V, the velocity component emitted from the test point and perpendicular to the second grid is

All secondary electrons larger than m/sec penetrate the second grid. More secondary electrons pass through the second grid when the potential at the test point is -2.5V than when it is 0V.
The amount of secondary electrons passing through the second grid is thus a function of the voltage at the test point and can be used to indicate or measure the voltage at the test point. For convenience, the above description omits consideration of changes in secondary emission depending on the sample material or the incident angle of the fine particles. This is because the normalization effect eliminates this variation. Secondary electrons (low energy) penetrating the second grid 26 are attracted toward the third grid 27, which is constituted by an annular network and is maintained at a positive voltage of 500V or more. The third grid 27 is similar to the first and second grids.
has high electron transmittance. Electrons penetrating the third grid 27
It is accelerated towards an annular plastic sinter 28 covered with a thin layer of aluminum 29 held at 10 KV. Most of the electrons penetrate the thin aluminum layer 29 and cause scintillation. Most of the scintillated light is collected by a bundle of optical fibers 30 bonded to the vertical outer surface of the annular scintillator 28. The optical fibers are bundled to deliver the received light to the cathode of a photomultiplier tube 31 (shown in Figure 2), whose output is amplified by an amplifier 32 having a transresistance amplifier providing an output voltage at a low output impedance. Ru. The voltage output displays the voltage at the test point on the printed wiring board. The electron beam that typically constitutes an electronic probe is stationary relative to each point being tested in order to properly ascertain the test voltage and to compare the signal with which the particular test point can be identified. Nevertheless, more than 1000 test points can be examined within 100 mS using the apparatus shown in Figure 1. FIG. 2 shows a control system with a minicomputer for directing the electronic probe 1. FIG. The control system mainly consists of a minicomputer 33,
It reads a test program from a storage device 34 adapted to receive data from a floppy disk 35 on which an appropriate program for testing a particular circuit has been previously recorded. This system uses amplifier 3
2, an accompanying logic circuit 37, and a digital/analog converter 38 that controls the deflection amplifier 39 of the electron gun.
and an interface device 36 for supplying control signals to an erase gate/amplifier 40 that controls a power supply 41 that supplies power to various parts of the electron gun. The control system also includes a power supply 42 for powering the focusing coil 5 and a power supply 43 for powering the various grids of the electronic collector, the power supply 43 being controlled by the minicomputer 33 via an interface device 36. has been done. The system also has a functional test device 44 controlled by the minicomputer 33. The test procedure begins by loading the circuit to be tested into the test chamber and loading a test program previously recorded on a floppy disk into the storage device 34. The test program includes a logical test sequence that performs testing and, if desired, fault diagnosis at optimal speed. A typical program includes both functional tests and probe tests. If the sequence specifies a functional test for which the use of an electronic probe is unnecessary or inappropriate, the functional test is performed by a functional test device, which performs a suitable test of the circuit to be tested in a known manner under the control of the minicomputer 33. Appropriate inputs are provided to the terminals and output signals obtained from the circuit under test are provided to the minicomputer 33. The electronic probe is erased when only functional testing is performed. However, most programs specify a combination of functional and probe tests. It is desirable to ensure that the electronic probe is accurately aligned with the pattern before such a test sequence begins. Customary work marks on printed circuit boards can be used for this purpose. These marks can be easily located by sweeping the beam over a large area relative to the mark and monitoring appropriate changes in the output signal of amplifier 32. The bias of the X, Y deflection coils can then be adjusted appropriately. With the aid of the minicomputer 33, the alignment process can easily be carried out automatically. Mini computer 3 for probe testing
3 generates the appropriate input signals to the test circuit with the aid of functional test equipment 44 and interface equipment 36 which transmits encoded signals specifying the coordinates (usually "X" and "Y" coordinates) of each point to be tested in turn. supply to Interface device 36 provides separate binary numbers to digital-to-analog converter 38 to generate X and Y deflection signals, which are transmitted to deflection amplifier 39.
is supplied to the deflection coil 6 by. The minicomputer 33 also causes the erase gate/amplifier 40 to turn on the electron probe 1 after an appropriate time for the deflection current to stabilize and then erase the electron beam before the probe moves to the next test point. The minicomputer 33 also sets the supply voltage to the second grid 26 to a specific voltage, such as a positive voltage of 10 V, during the first part of the rest time of the electronic probe 1 on the test point. For this particular voltage of the second grid 26, the output voltage V _co of the amplifier 32 is characteristic of the summed or normalized low energy secondary electron emission. During the final portion of the quiescent time, the voltage on the second grid 26 is switched to the filter voltage value (between 0 and -5V as described above) and the amplifier 32 is switched during this final portion of the quiescent time.
The output voltage V _cv indicates the voltage characteristics of the test point. Here, the secondary electron beam generated by the collision of the primary electron beam with the electronic circuit network is emitted at different speeds and trajectories depending on the irradiation angle of the primary electron beam.
Therefore, the number of secondary electrons passing through a single region at a single time varies depending on the angle that the path of the secondary electron beam makes with the path of the primary electron beam. This angle-dependent change in the number of secondary electrons is called the angular distribution of secondary emission.
Further, if there is a change in the vertical position of the test point, the value will change due to this change. Therefore, for each test point, if we take the ratio of the normalized output voltage V _co at the start of the rest time to the output voltage V _cv at the end of the rest time, we can calculate the test voltage from the test point voltage. The effects of point position variations and angular distributions can be canceled out, and the ratio indicates the ratio of the test point voltage to the normal voltage. Therefore, by multiplying this ratio by a predetermined coefficient, the actual voltage value at the test point can be obtained. Finally, interface device 36 opens the gates in the logic circuit of head amplifier 37 for a precise period of time and records the values of V _co and V _cv . The ratio of these two values is compared within the logic circuit of the head amplifier with the "correct" value supplied by the minicomputer 33 to the comparator. The comparison result is minicomputer 33
Feedback is received to determine whether to proceed with the next test in the sequence. The minicomputer 33 moves the electronic probe 1 through designated test points for a given combination of input signals and the values are recorded to compare the full electronic circuit with the correct full electronic circuit or to test specific gates or Faults within specific subsystems of an electronic circuit can be independently indicated. The minicomputer 33 can easily be programmed to perform complete fault diagnosis and display the nature and location of the fault by output to a visual display or other peripheral device. Using an electronic collector sensitive to electromagnetic fields,
It is generally possible to test electron flow in conductors and circuit components rather than voltage at connection points. In general, large-scale testing involves testing all forms of electronic circuits, such as medium-sized integrated circuits, individual components, microwave integrated circuits, and surface acoustic wave devices with metal conductors mounted on piezoelectric materials, using the same method as above. Can be done. As described above, in the present invention, the electric potential at each point is measured by temporarily stopping the electronic probe at each of a series of points in the electronic circuit network to be tested and measuring the secondary electron flow at that time. , it is possible to measure more complex printed circuits and integrated circuits faster than measurements with mechanical probes, and in addition, during this rest period, a potential barrier filter is applied to the secondary electron flow, and the filter potential is switched during the rest period. Since a signal representing the potential at the test point is obtained from the ratio between the signals obtained, changes in the position of the test point and minute changes in the incident angle of the electron beam can be normalized, and the voltage at the test point can always be measured correctly.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram of a scanning device, and FIG. 2 is a schematic diagram of the scanning device in conjunction with a control system. Explanation of reference symbols, 1...electronic probe, 2...
Electron gun, 5... Focusing coil, 6... Deflection coil,
23...first grid, 25...printed wiring board, 2
6...Second grid, 28...Scintillator, 3
2...Head amplifier, 33...Mini computer, 34...Storage device, 35...Floppy disk, 36...Interface device, 37...Logic circuit, 38...Digital/analog converter, 39
... deflection amplifier, 40 ... erasure gate/amplifier,
41...Electron gun power supply, 42...Focusing coil power supply, 43...Collector power supply, 44...Function test device.

Claims (1)

[Scope of Claims] 1. A method for testing the operation of a completed electronic network having accessible wiring that requires special excitation during operation, comprising: applying a corresponding signal to the wiring of said electronic network; () directing an electron probe in an orderly sequence with an electron beam to a programmed series of points within said electronic network; and () moving said electronic probe. for each of the series of points in the electronic network; () applying a potential barrier filter to the secondary electron flow caused by the collision of the electron beam during the pause time for each of the series of points; allows a secondary electron flow to pass through according to the potential difference between each point and a predetermined filter potential and outputs this electron flow as a signal; () the potential barrier filter is disposed within the path of the electron probe; in the form of two electrically transparent grids that are connected and out of focus with respect to the electronic probe, and a high positive potential is applied to the grid that is closer to the electronic network. and applying the predetermined filter potential to the grid further from the electronic circuit network; () switching the predetermined filter potential from a relatively high value to a negative value during the rest time; A method for testing an electronic circuit network, in which the ratio of the values of both output signals is calculated, and the value represents the potential at a point on the electronic circuit network.